Secondary Logo

Journal Logo

Observational Study

Social Determinants of Health and Delirium Occurrence and Duration in Critically Ill Adults

Wu, Ting-Ting PharmD1,2; Zegers, Marieke PhD3; Kooken, Rens BSc3; Griffith, John L. PhD4; Molnar, Beth E. ScD4; Devlin, John W. PharmD, MCCM1,5; van den Boogaard, Mark RN, PhD3

Author Information
doi: 10.1097/CCE.0000000000000532
  • Open

Abstract

Social determinants of health (SDOH) are categorized by the Centers for Disease Control into five domains: social/community context, education access/quality, economic stability, neighborhood/built environment, and healthcare access/quality (1). The first three domains are often grouped as socioeconomic factors. There is great interest in determining the influence of SDOH on ICU outcomes given social policy changes may sometimes help improve SDOH (2). The results of published studies investigating this question are mixed in terms of whether a relationship between SDOH and critical care outcomes exists (2–5). One investigation in the Nashville, TN, area found that patients who are non-White and have a lower education are more likely to experience greater post-ICU, long-term cognitive impairment (3). However, in one Australian state, a relationship between socioeconomic status and ICU mortality was not demonstrated (4). A French study found no relationship between socioeconomic status and mortality or psychologic health 1 year after ICU discharge (5).

Delirium, a common sequela of critical illness, is associated with poor outcomes (6,7). Multiple predisposing and precipitating factors increase ICU delirium risk (7,8). A longer ICU delirium duration increases mortality and worsens long-term cognitive impairment (7,9). Although African Americans (vs Caucasians) have a similar risk for delirium in the ICU (10), the association between other SDOH and delirium has not been evaluated. Ecosocial theory integrates social and biologic characteristics and suggests that adults who have been chronically exposed to the disadvantages conferred by SDOH may have prolonged immune activation resulting in physiologic dysregulation and excessive inflammatory response (11). With delirium being an acute neuroinflammatory condition (6), we hypothesized critically ill adults having chronically poor SDOH are at greater risk for ICU delirium. We evaluated the associations between three SDOH domains and the occurrence and duration of delirium in critically ill adults.

MATERIALS AND METHODS

This retrospective cohort study is a secondary, subgroup analysis of the Monitoring cOnsequeNces of InTensive care fOR Intensive Care patients (MONITOR-IC) study (ClinicalTrials.gov NCT03246334) (12,13). The multicenter, MONITOR-IC cohort study prospectively compared patient self-reported cognitive, psychologic, and physical health statuses before- and 1-year after ICU admission in 2,345 critically ill adults in the Netherlands (12,13). Consecutive patients enrolled in the MONITOR-IC study between July 2016 and February 2020 who were admitted to the 36-bed mixed medical-surgical ICU at Radboud University Medical Center (RadboudUMC), an academic medical center in Nijmegen, NL, were included in this study. Patients admitted after elective surgery, who were currently residing in a long-term care facility, or not expected to survive greater than or equal to 48 hours were excluded. This study was approved by the ethics committee of the RadboudUMC (2016-2724). Each participant, or their legal representative, provided written informed consent.

Baseline demographic and SDOH data were obtained from the MONITOR-IC database (12,13); all remaining patient data were extracted from the RadboudUMC EPIC (Verona, WI) electronic health record. Data on four SDOH were collected via questionnaires filled in by patients (or families) at the time of ICU admission: 1) social/community context (ethnicity: non-Dutch vs Dutch [patient and both parents NL-born]), 2) education access/quality (highest educational attainment: low [≤ secondary school graduate] vs high [postsecondary graduate]), and 3) economic stability (employment status: part-time employed/unemployed vs fully employed and individual monthly income: low [≤ $USD 2,075] vs high [> $3,052]) (1,12). The monthly income levels chosen were consistent with the income levels (in EUROS) used in the MONITOR-IC trial (12). Patients greater than or equal to 65 years were assumed not to be part of the active workforce and were excluded from the employment status analysis.

Bedside ICU nurses evaluated patients without coma (Richmond Agitation Sedation Scale [RASS], –4 or –5) (14) every 8 hours with the Confusion Assessment Method-ICU (CAM-ICU) when they were maximally awake (e.g., after a spontaneous awakening trial) (15,16). Both RASS and CAM-ICU assessment are longstanding standards of care in the study ICU; nurses undergo regular training updates. A delirium day was defined by greater than or equal to 1 + CAM-ICU and/or scheduled antipsychotic use (16). Data on ICU days and 10 variables previously shown to be associated with increased ICU delirium or duration (7,8) were collected: before-ICU: age, modified Charlson comorbidity index (mCCI) (17), cognitive impairment (Cognitive Failures Questionnaire Score ≥ 43) (18), antidepressant, antipsychotic, or benzodiazepine use (indicative of a preexisting mental illness); ICU baseline: admission type and Acute Physiology and Chronic Health Evaluation IV score (19); and Daily ICU (until delirium occurrence or discharge): average Sequential Organ Failure Assessment score (20), restraint use, coma occurrence, daily ICU benzodiazepine (≥ 5-mg midazolam equivalent) (21), and opioid (≥ 10-mg IV morphine equivalents) use (22).

Logistic regression models were created to measure the association between each of the four SDOH and ICU delirium occurrences. Quasi-Poisson regression models were created to measure the association between each of the four SDOH and ICU delirium durations. Patients with random missing data (SDOH [ethnicity 6.6%, education 3.3%, employment status 4.7%, and income 14.7%] and covariates [mCCI 4.8%]) were excluded from analyses. All models were adjusted for all pre-ICU, ICU baseline, and daily ICU covariates. Significance was defined as a two-sided p value of less than 0.05. All analyses were performed using R version 4.0.3 (R Foundation for Statistical Computing).

RESULTS

Among 956 patients, 9.4% were non-Dutch, 75.3% had low educational attainment, 35.9% (among patients ≤ 65 years old) were unemployed or worked only part-time, and 51.2% earned a low monthly income (Table 1). Delirium occurred in 393/956 of the patients (45.4%) for a median (interquartile range) of 2 days (1–5 d). Low-income patients were more likely to experience ICU delirium than high-income patients (p = 0.05) (Table 1). All covariates were significantly different between the delirium and nondelirium groups except baseline cognitive impairment, baseline benzodiazepine use, and admission type (Table 2).

TABLE 1. - Comparison of Select Social Determinant of Health Between Patients With and Without ICU Delirium
Social Determinant All Patients All Patients p
Delirium No Delirium
Social and community context
 Ethnic background, n (%)
 Total 893 (100) 361 (40) 532 (60) 0.13
 Non-Dutch 84 (9) 41 (49) 43 (51)
 Dutch 809 (91) 320 (40) 489 (60)
Education access and quality
 Educational attainment, n (%)
  Total 924 (100) 385 (42) 539 (58) 0.82
  Low 696 (75) 292 (42) 404 (58)
  High 228 (25) 93 (41) 135 (59)
Economic stability
 Employment status, n (%)
  Total 537 (100) 195 (36) 342 (64) 0.86
  Part-time employed/unemployed 193 (36) 71 (37) 122 (63)
  Full-time employment 344 (64) 124 (36) 220 (64)
 $ U.S. monthly income, n (%)
  Total 815 (100) 325 (40) 490 (60)
  Low 417 (51) 180 (43) 237 (57) 0.05
  Medium 239 (29) 89 (37) 150 (63) 0.32
  High 159 (20) 56 (35) 103 (65) 0.18

TABLE 2. - Comparison of Model Covariables Between Patients With and Without ICU Delirium
Model Covariable All Patients (n = 956) All Patients p
Delirium (n = 393) No Delirium (n = 563)
Pre-ICU admission variables
 Age, median (IQR) 63 (51–71) 65 (54–72) 61 (49–70) < 0.01
 Modified Charlson comorbidity index, median (IQR) 2 (1–4) 3 (2–4) 2 (1–4) < 0.01
 Cognitive impairment, n (%) 42 (4) 21 (0.5) 21 (0.4) 0.23
 Scheduled antipsychotic, antidepressant or benzodiazepine   use, n (%) 35 (4) 22 (0.6) 13 (0.2) < 0.01
ICU baseline variables
 Admission type, n (%)
  Medical 345 (36) 250 (41) 361 (59) 0.93
  Surgical 611 (64) 143 (41) 202 (59)
 APACHE IV score, median (IQR) 67 (52–83) 75 (61–93) 61 (47–77) < 0.01
Daily ICU variables
 Use of physical restraint, n (%) 593 (62) 313 (53) 280 (47) < 0.01
 Sequential Organ Failure Assessment score, median (IQR) 6 (4–8) 8.0 (6–10) 4.8 (3–7) < 0.01
 Presence of coma, n (%) 347 (36) 226 (65) 121 (35) < 0.01
 Opioid use, n (%) 278 (29) 274 (99) 4 (1) < 0.01
 Benzodiazepine use, n (%) 178 (19) 176 (99) 2 (1) < 0.01
 Days spent in the ICU, median (IQR) 2 (1–3) 1 (0–4) 2 (1–3) < 0.01
IQR = interquartile range.

Across the multivariate models, none of the four SDOH was significantly associated with increased delirium occurrence or duration (Table 3). Low (vs high) monthly income was weakly associated with increased delirium occurrence (adjusted odds ratio, 1.83; 95% CI, 0.91–3.89). Low (vs high) educational attainment (adjusted relative risk, 1.21; 95% CI, 0.97–1.53) was weakly association with more ICU delirium days.

TABLE 3. - Association Between Social Determinant of Health and ICU Delirium Occurrence and Duration
Social Determinant Delirium Occurrence Delirium Duration
Adjusted OR (95% CI) p Attributable Risk Ratio (95% CI) p
Ethnicity (n = 849)
 Dutch (R) 1 1
 Non-Dutch 1.04 (0.41–2.42) 0.94 1.27 (0.94–1.69) 0.11
Educational attainment (n = 881)
 High (R) 1 1
 Low 0.80 (0.47–1.39) 0.43 1.21 (0.97–1.53) 0.10
Employment status (n = 516)
 Full-time employment (R) 1 1
 Part-time employment or unemployed 0.95 (0.42–2.09) 0.91 0.96 (0.71–1.30) 0.80
Monthly income (n = 778)
 High (R) 1 1
 Medium 1.08 (0.48–2.47) 0.85 1.15 (0.85–1.57) 0.36
 Low 1.83 (0.91–3.89) 0.10 1.15 (0.88–1.52) 0.30
OR = odds ratio, R = reference.

DISCUSSION

The potential relationships between SDOH and the adverse outcomes associated with critical illness have been at the forefront of the current severe acute respiratory coronavirus 2 (SARS-CoV-2) pandemic (23,24). Delirium is highly prevalent during critical illness and associated with deleterious ICU and post-ICU outcomes (7,9). Recognition of delirium risk factors, particularly if they are modifiable, is a key component of ICU delirium reduction efforts (7,8). Our study is the first publication to explore the association between SDOH and ICU delirium occurrence and duration. If societal improvements (e.g., better access to quality education and better employment/higher income) lower chronic inflammation over the lifespan, the risk for delirium may be reduced in populations at high risk for developing it (e.g., the critically ill) (2,6,11).

Although we were unable to find evidence from one Dutch region that SDOH affect ICU delirium occurrence or duration, one should not assume these relationships do not exist in other populations. By evaluating patients from one relatively homogeneous region, our results may not apply to other countries or settings where SDOH may be quite different (1,2). We did not consider data on race and healthcare access/quality in our study because the Nijmegen, NL, region is primarily White, and all Dutch citizens have access to high-quality medical care (1,2). SDOH variability may exist between individual neighborhoods in the Nijmegen region, but the Dutch government does not allow postal codes to be used in research.

Future research exploring the relationship between SDOH and ICU delirium is required in critically ill populations having diverse SDOH, despite Australian and French studies not demonstrating an association between SDOH and mortality or post-ICU psychologic health (4,5). Non-White race and low educational attainment have each been shown to be associated with greater long-term cognitive impairment after ICU discharge for patients hospitalized in Nashville, TN—a region where SDOH epidemiology may be different from Australia or Europe (3,5). Although Haddad et al (3) did not consider ICU delirium occurrence or duration in their Nashville-based study, delirium is strongly associated with greater long-term cognitive impairment (9). In older adult Americans admitted to the floor after major surgery, studies demonstrate low education and income level are associated with greater postoperative delirium (25,26). The authors of these reports hypothesized lower education may reduce cognitive reserve and thus increase delirium (25,26). With level of education and level of income closely interrelated, it remained unclear in either report whether low-income alone was an independent risk for postoperative delirium.

Our study has important strengths. We evaluated close to 1,000 critically ill adults, prospectively collected data on factors related to three SDOH domains, rigorously evaluated delirium three times daily, and accounted for 10 established ICU delirium risk factors in all analyses. Our study also has limitations. By excluding non-Dutch-speaking patients (because of the challenge of screening them for delirium with the CAM-ICU), we may have decreased the number of patients with a non-Dutch ethnicity in our cohort. Response or recall bias may have occurred during SDOH questionnaire completion. Validated socioeconomic scores were not used (3,4). Although all ICU variables were collected before delirium first occurred, formal time-dependent analyses were not used, and thus, residual confounding may have resulted.

CONCLUSION

In conclusion, our study suggests SDOH may not affect ICU delirium occurrence or duration in one relatively homogenous Dutch region. Further research in diverse regions of the world, where multiple factors across all five SDOH domains can be evaluated, needs to be completed in critically ill adults, both with and without coronavirus disease 2019, to better define the association between SDOH and delirium in critically ill adults.

REFERENCES

1. Centers for Disease Control and Prevention. Social Determinants of Health: Know What Affects Health. Available at: https://www.cdc.gov/socialdeterminants. Accessed June 26, 2021
2. Jones JRA, Berney S, Connolly B, et al. Socioeconomic position and health outcomes following critical illness: A systematic review. Crit Care Med. 2019; 47:e512–e521
3. Haddad DN, Mart MF, Wang L, et al. Socioeconomic factors and intensive care unit-related cognitive impairment. Ann Surg. 2020; 272:596–602
4. Mullany DV, Pilcher DV, Dobson AJ. Associations between socioeconomic status, patient risk, and short-term intensive care outcomes. Crit Care Med. 2021; 49:e849–e859
5. Bastian K, Hollinger A, Mebazaa A, et al.; FROG-ICU Study Investigators. Association of social deprivation with 1-year outcome of ICU survivors: Results from the FROG-ICU study. Intensive Care Med. 2018; 44:2025–2037
6. Wilson JE, Mart MF, Cunningham C, et al. Delirium. Nat Rev Dis Primers. 2020; 6:90
7. Devlin JW, Skrobik Y, Gélinas C, et al. Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit Care Med. 2018; 46:e825–e873
8. Zaal IJ, Devlin JW, Peelen LM, et al. A systematic review of risk factors for delirium in the ICU. Crit Care Med. 2015; 43:40–47
9. Pandharipande PP, Girard TD, Jackson JC, et al.; BRAIN-ICU Study Investigators. Long-term cognitive impairment after critical illness. N Engl J Med. 2013; 369:1306–1316
10. Khan BA, Perkins A, Hui SL, et al. Relationship between African-American race and delirium in the ICU. Crit Care Med. 2016; 44:1727–1734
11. Krieger N. Methods for the scientific study of discrimination and health: An ecosocial approach. Am J Public Health. 2012; 102:936–944
12. Geense W, Zegers M, Vermeulen H, et al. MONITOR-IC study, a mixed methods prospective multicentre controlled cohort study assessing 5-year outcomes of ICU survivors and related healthcare costs: A study protocol. BMJ Open. 2017; 7:e018006
13. Geense WW, Zegers M, Peters MAA, et al. New physical, mental, and cognitive problems 1 year after ICU admission: A prospective multicenter study. Am J Respir Crit Care Med. 2021; 203:1512–1521
14. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: Validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002; 166:1338–1344
15. Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill patients: Validation of the confusion assessment method for the intensive care unit (CAM-ICU). Crit Care Med. 2001; 29:1370–1379
16. van den Boogaard M, Slooter AJC, Brüggemann RJM, et al.; REDUCE Study Investigators. Effect of haloperidol on survival among critically ill adults with a high risk of delirium: The REDUCE randomized clinical trial. JAMA. 2018; 319:680–690
17. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 1987; 40:373–383
18. Broadbent DE, Cooper PF, FitzGerald P, et al. The cognitive failures questionnaire (CFQ) and its correlates. Br J Clin Psychol. 1982; 21:1–16
19. Zimmerman JE, Kramer AA, McNair DS, et al. Acute physiology and chronic health evaluation (APACHE) IV: Hospital mortality assessment for today’s critically ill patients. Crit Care Med. 2006; 34:1297–1310
20. Vincent JL, de Mendonça A, Cantraine F, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: Results of a multicenter, prospective study. Working group on “sepsis-related problems” of the European Society of Intensive Care Medicine. Crit Care Med. 1998; 26:1793–1800
21. Zaal IJ, Devlin JW, Hazelbag M, et al. Benzodiazepine-associated delirium in critically ill adults. Intensive Care Med. 2015; 41:2130–2137
22. Duprey MS, Dijkstra-Kersten SMA, Zaal IJ, et al. Opioid use increases the risk of delirium in critically ill adults independently of pain. Am J Respir Crit Care Med. 2021 Apr 9. [online ahead of print]
23. Abrams EM, Szefler SJ. COVID-19 and the impact of social determinants of health. Lancet Respir Med. 2020; 8:659–661
24. Singu S, Acharya A, Challagundla K, et al. Impact of social determinants of health on the emerging COVID-19 pandemic in the United States. Front Public Health. 2020; 8:406
25. Arias F, Chen F, Fong TG, et al. Neighborhood-level social disadvantage and risk off delirium following major surgery. J Am Ger Soc. 2020; 68:2863–2871
26. Martins S, Paiva JA, Simões MR, et al. Delirium in elderly patients: Association with educational attainment. Acta Neuropsychiatr. 2017; 29:95–101
Keywords:

delirium; intensive care; social determinants of health; socioeconomic factors

Copyright © 2021 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of the Society of Critical Care Medicine.