Impact of Preexisting Dementia on Mortality, Readmissions, and Complications After Surgical Repair of Hip Fractures: A Cohort Analysis Authors: Spencer T. Rasmussen, Kamal Shaik, Kimberly Dong, Zachary Demetriou, Yale A. Fillingham, Asif M. Ilyas Published: 2026 ABSTRACT Objectives:As the population ages, hip fractures are projected to increase. These fractures carry high morbidity and mortality in older adults. Dementia, common among these patients, may further worsen outcomes. This study evaluates the impact of preexisting dementia on mortality after surgical repair of femoral neck and intertrochanteric fractures, with the study hypothesis that the presence of preoperative dementia can result in increased mortality postoperatively.Methods:Using the TriNetX Research Network, 68,639 patients aged 65 to 90 years who underwent surgical treatment of femoral neck or intertrochanteric fractures from 2015 to 2025 were identified. Patients were grouped based on the presence or absence of a dementia diagnosis before fracture. Surgeries included hemiarthroplasty, internal fixation, or intramedullary nailing. Propensity score matching adjusted for age, sex, and comorbidities. The primary outcome was all-cause mortality at 90 days, 2, and 5 years for patients who had sufficient follow-up data. Secondary outcomes included 90-day readmissions and complications (pneumonia, stroke, urinary tract infection, and heart failure). Risk ratios (RRs) with 95% confidence intervals (CIs) were calculated.Results:Patients with dementia had higher mortality at all time points. In intertrochanteric fractures, dementia patients had increased mortality at 90 days (RR, 0.61; 95% CI, 0.57 to 0.65; P < 0.001), 2 years (RR, 0.68; 95% CI, 0.65 to 0.70; P < 0.001), and 5 years (RR, 0.76; 95% CI, 0.73 to 0.78; P < 0.001). In femoral neck fractures, mortality was also higher at 90 days (RR, 0.56; 95% CI, 0.49 to 0.63; P < 0.001), 2 years (RR, 0.64; 95% CI, 0.59 to 0.70; P < 0.001), and 5 years (RR, 0.71; 95% CI, 0.67 to 0.76; P < 0.001) in patients with dementia. At 90 days for both fracture types, the risk of mortality was nearly double among dementia patients. Similarly, 90-day readmissions and postoperative complications, including pneumonia, stroke, urinary tract infection, heart failure, and revision surgery, were statistically significantly higher among dementia patients.Conclusion:Preexisting dementia is associated with higher mortality after hip fracture repair in older adults, approaching nearly double at 90 days. Readmissions and complications are also markedly higher among dementia patients. These findings underscore the importance of tailored perioperative planning, enhanced care pathways for patients with cognitive impairment, and shared decision making between the patient, caregivers, and the care teams. Hip fractures have long been recognized as a notable public health concern, representing a leading cause of mortality, disability, and healthcare expenditure among the elderly population.1-4 Numerous national and international epidemiological studies have projected a rising incidence of hip fractures driven by global population aging.5-8 Most hip fractures in older adults result from falls, often precipitated by balance and gait impairments associated with aging and neurocognitive disorders such as dementia.9,10 Dementia is also recognized as an independent predictor of increased postoperative mortality in patients undergoing hip fracture repair.11-13 Prior studies have examined the influence of preexisting dementia on postoperative outcomes in hip fracture patients, although typically limited to single-center analyses. To date, no studies, to our knowledge, have leveraged large multicenter databases to examine this relationship while stratifying by intertrochanteric and femoral neck fracture types. Therefore, this study uses the TriNetX Research Network to evaluate the impact of preexisting dementia on postoperative mortality in older adults undergoing surgical repair of femoral neck and intertrochanteric hip fractures. The study hypothesis was that patients with preexisting dementia would be at an increased risk of mortality. Methods The TriNetX Research Network, a global, federated health research platform, was queried retrospectively on April 29, 2025. The TriNetX database does not involve patient identifiable information and is subsequently exempt from institutional review board review and approval. TriNetX is a global research network encompassing data from more than 170 Healthcare Organizations (HCOs) and more than 400 million patients.14 It contains deidentified, aggregate patient information covering procedures, diagnoses, medications, vital signs, genomics, and demographics. HCOs involved in the TriNetX network contribute healthcare data in deidentified, pseudoanonymized, or limited data set formats, in accordance with local privacy regulations. These HCOs authorize the usage of these data for research purposes on the TriNetX platform. In return for providing data, HCOs incur no financial expenses and gain access to data query tools, analytics, visualization capabilities, and the necessary hardware for software execution. The deidentification process conforms to the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule standards, as verified by a qualified expert, meeting the requirements of Section §164.514(b)(1), ensuring HIPAA compliance. Four total cohorts containing patients aged 65 to 90 years between the years January 1, 2015, and January 1, 2025, were identified using International Classification for Diseases (ICD) 10th edition and Current Procedural Terminology (CPT) codes as inclusion and exclusion criteria. Initially, patients were identified based on the presence of an intertrochanteric fracture (ICD—S72.1) or femoral neck fracture (S72.0) on or before surgical repair of these respective fractures (intertrochanteric fracture: open reduction and internal fixation—CPT 27236 or intramedullary nailing—CPT 27245; femoral neck fracture: hemiarthroplasty—CPT 27125 or percutaneous pinning—CPT 27235). Total hip arthroplasty (THA; CPT-27130) was omitted from the femoral neck fracture cohort. Although THA is an increasingly common treatment option, its clinical management differs from hemiarthroplasty, internal fixation, or intramedullary nailing. Therefore, THA was excluded to minimize variability and avoid procedural bias within the cohorts. The study cohort then consisted of patients with a preoperative history of a dementia diagnosis (Alzheimer disease [G30], senile degeneration of the brain [G31.1], dementia with Lewy bodies [G31.83], vascular dementia [F01], dementia in other diseases [F02], or unspecified dementia [F03]) before surgical treatment. The control cohort consisted of patients without a preoperative history of dementia. Propensity score matching for age, sex, race, ethnicity, body mass index (ICD—Z68.1 to Z68.4), and other comorbidities (ie, diabetes mellitus—E08 to E13; primary hypertension—I10; heart failure—I50; chronic ischemic heart disease—I25; atrial fibrillation—I48; overweight or obesity—E65 to E68; chronic kidney disease—N18; chronic obstructive pulmonary disease—J44; and osteoporosis—M81) were used to balance baseline characteristics between dementia and nondementia cohorts using a 1:1 nearest-neighbor matching algorithm without replacement. All-cause mortality at 90 days, 2, and 5 years was the primary outcome of interest. Secondary postoperative outcomes included readmission (CPT—1013659) between 5 and 90 days, as well as joint infection (ICD—T84.51 to T84.52), pneumonia (ICD—J12 to J18), cardiovascular events (ICD—I21 to I24), stroke (ICD—I60 to I63), urinary tract infection (UTI; ICD—N10, N30, N39.0), revision arthroplasty (CPT—27132), nonunion (ICD—S72.141K, S72.142K), malunion (ICD—S72.141P, S72.142P), heart failure (ICD—I50), and respiratory complications (intraoperative respiratory complications [J95] or respiratory failure [J96]) at 90 days. Patients within the initial propensity score–matched cohorts were included in the longitudinal analysis to determine the observed cumulative incidence of each outcome. Statistical analyses were run through the TriNetX platform. A P value of 0.05 was used to indicate statistical significance. Cohort demographic and comorbid differences were assessed using two-sided independent sample t-tests and chi-squared tests. Match quality was evaluated using P values and standardized mean differences. Risk ratios (RRs) with 95% confidence intervals (CIs) were calculated to compare primary and secondary outcomes between cohorts within TriNetX, and then summary statistics were manually extracted. Results A total of 68,629 patients who sustained a hip fracture that was treated surgically were identified using the inclusion and exclusion criteria before propensity score matching. Matched characteristics between the nondementia (n = 14938) and dementia (n = 14938) intertrochanteric fracture groups are presented in Table 1. The nondementia group had a greater number of White patients (P = 0.0116) and a lower number of black or African American patients (P = 0.0414), both of which yielded statistically significant differences despite making up similar percentages of the cohorts. A greater number of patients in the dementia group had a body mass index of less than 19.9 kg/m2 (P = 0.0151). Table 1Matched Patient Characteristics—Intertrochanteric Fracture Repair CohortsIntertrochanteric Fracture RepairCharacteristicAfter Matching Sample No. (%)No Dementia (n = 14938)Dementia (n = 14938)P ValueSMDAge at index (yr)84 ± 784 ± 6Demographics Male3790 (25%)3821 (26%)0.68060.0048 Female10228 (68%)10231 (68%)0.97020.0004 White12147 (81%)11975 (80%) 0.0116 0.0292 Black or African American653 (4%)727 (5%) 0.0414 0.0236 Asian532 (4%)569 (4%)0.25590.0131 Hispanic or Latino553 (4%)607 (4%)0.10580.0187 Not Hispanic or Latino11374 (76%)11363 (76%)0.88140.0017 Unknown ethnicity3010 (20%)2967 (20%)0.53400.0072Comorbidities Essential (primary) hypertension11763 (79%)11696 (78%)0.34520.0109 Chronic ischemic heart disease5577 (37%)5523 (37%)0.51790.0075 Osteoporosis without current pathological fracture4891 (33%)4858 (33%)0.68390.0047 Atrial fibrillation and flutter4751 (32%)4754 (32%)0.97030.0004 CKD4786 (32%)4766 (32%)0.80400.0029 Diabetes mellitus4475 (30%)4502 (30%)0.73330.0039 Heart failure4281 (29%)4340 (29%)0.45120.0087 Other COPD2932 (20%)2989 (20%)0.40810.0096 Overweight, obesity, and other hyperalimentation1742 (12%)1802 (12%)0.28300.0124BMI (kg/m2) ≤19.91186 (8%)1302 (9%) 0.0151 0.0281 20-292136 (14%)2210 (15%)0.22460.0141 30-39931 (6%)951 (6%)0.63390.0055 ≥40167 (1%)181 (1%)0.45030.0087Bold entries are significant p-value.BMI = body mass index, CKD = chronic kidney disease, COPD = chronic obstructive pulmonary disease, SMD = standardized mean difference Matched Patient Characteristics—Intertrochanteric Fracture Repair Cohorts Bold entries are significant p-value. BMI = body mass index, CKD = chronic kidney disease, COPD = chronic obstructive pulmonary disease, SMD = standardized mean difference All-cause mortality in the intertrochanteric fracture repair dementia group was markedly higher than their nondementia counterparts at all postoperative time points (90-days: RR = 0.61, 95% CI [0.57 to 0.65], P < 0.0001; 2-years: RR = 0.68, 95% CI [0.65 to 0.70], P < 0.0001; 5-years: RR = 0.76 [0.73 to 0.78], P < 0.0001; Figure 1). At 90 days, the risk of mortality was nearly double among dementia patients. Although this increased risk lessened at 2 and 5 years, it remained statistically significantly higher at all time points. In addition, patients with dementia were more likely to be readmitted between 5 and 90 days (RR = 0.95, 95% CI [0.92 to 0.97], P < 0.0001), contract pneumonia (RR = 0.88, 95% CI [0.79, 0.99], P = 0.0286), sustain a stroke (RR = 0.75, 95% CI [0.63 to 0.88], P = 0.0005), and contract a UTI (RR = 0.80, 95% CI [0.74 to 0.87], P < 0.0001) at 90 days postoperatively. Patients without dementia were more likely to have a revision arthroplasty (RR = 2.18, 95% CI [1.45 to 3.27], P = 0.0001) and be diagnosed with heart failure (RR = 1.14, 95% CI [1.02 to 1.27], P = 0.0225) at 90 days postoperatively (Table 2). No notable differences were observed between the groups in the rates of joint infection, cardiovascular events, nonunion, malunion, or respiratory complications. Bar graph showing intertrochanteric fracture repair mortality risk. Table 2Intertrochanteric Fracture Repair Secondary 90-Day OutcomesNo Dementia (n = 14938)(% Risk)Dementia (n = 14938)(% Risk)RR (95% CI)P Value90-d outcomes Readmission6411 (45%)6779 (47%)0.95 (0.92-0.97) <0.0001 Joint infection44 (0.3%)44 (0.3%)1.00 (0.66-1.52)0.9990 Pneumonia539 (4%)582 (5%)0.88 (0.79-0.99) 0.0286 Cardiovascular events390 (3%)379 (3%)1.02 (0.89-1.18)0.7287 Stroke246 (2%)308 (3%)0.75 (0.63-0.88) 0.0005 UTI955 (10%)968 (12%)0.80 (0.74-0.87) <0.0001 Revision arthroplasty74 (0.5%)34 (0.2%)2.18 (1.45-3.27) 0.0001 Nonunion65 (0.4%)49 (0.3%)1.33 (0.92-1.92)0.1326 Malunion35 (0.2%)28 (0.2%)1.25 (0.76-2.05)0.3770 Heart failure638 (6%)558 (5%)1.14 (1.02-1.27) 0.0225 Respiratory complications749 (6%)753 (6%)0.97 (0.88-1.07)0.5647Bold entries are significant p-value.CI = confidence interval, RR = risk ratio, UTI = urinary tract infection Intertrochanteric Fracture Repair Secondary 90-Day Outcomes Bold entries are significant p-value. CI = confidence interval, RR = risk ratio, UTI = urinary tract infection Matched characteristics between the nondementia (n = 3863) and the dementia (n = 3863) femoral neck fracture groups are found in Table 3. All-cause mortality in the femoral neck fracture repair dementia group was markedly higher than their nondementia counterparts at all postoperative time points (90 days: RR = 0.56, 95% CI [0.49 to 0.63], P < 0.0001; 2 years: RR = 0.64, 95% CI [0.59 to 0.70], P < 0.0001; 5-years: RR = 0.71, 95% CI [0.67 to 0.76], P < 0.0001; Figure 2). At 90 days, the mortality risk was nearly twice as high among patients with dementia. Although this elevated risk diminished at 2 and 5 years, it remained statistically significantly higher at all evaluated time points. In addition, at 90 days postoperatively, patients with dementia were more likely to be readmitted (RR = 0.83, 95% CI [0.78 to 0.88], P < 0.0001), develop pneumonia (RR = 0.78, 95% CI [0.63 to 0.95], P = 0.0153), sustain a stroke (RR = 0.72, 95% CI [0.54 to 0.96], P = 0.0229), and develop an UTI (RR = 0.71, 95% CI [0.61 to 0.83], P < 0.0001) compared with those without dementia. Conversely, patients without dementia were markedly more likely to undergo revision arthroplasty (RR = 1.69, 95% CI [1.04 to 2.74], 0.0308) and be diagnosed with heart failure (RR = 1.28, 95% CI [1.04 to 1.59], P = 0.0223) during the same postoperative period (Table 4). No notable differences were observed between groups in the rates of joint infection, cardiovascular events, nonunion, malunion, or respiratory complications. Table 3Matched Patient Characteristics—Femoral Neck Fracture Repair CohortsFemoral Neck Fracture RepairCharacteristicAfter Matching Sample No. (%)No Dementia (n = 3863)Dementia (n = 3863)P ValueSMDAge at index (yr)83 ± 783 ± 7Demographics Male1144 (30%)1150 (30%)0.88120.0034 Female2621 (68%)2590 (67%)0.45160.0171 White3076 (80%)3006 (78%)0.05170.0443 Black or African American268 (5%)285 (7%)0.45310.0171 Asian197 (5%)211 (5%)0.47640.0162 Hispanic or Latino135 (4%)142 (4%)0.66840.0097 Not Hispanic or Latino3071 (80%)3039 (79%)0.37070.0204 Unknown ethnicity657 (17%)682 (18%)0.45240.0171Comorbidities Essential (primary) hypertension2924 (76%)2905 (75%)0.61550.0114 Chronic ischemic heart disease1356 (35%)1325 (34%)0.45880.0169 Osteoporosis without current pathological fracture1049 (27%)1042 (27%)0.85770.0041 Atrial fibrillation and flutter1127 (29%)1137 (29%)0.80260.0057 CKD1065 (28%)1089 (28%)0.54260.0139 Diabetes mellitus1040 (27%)1062 (27%)0.57380.0128 Heart failure969 (25%)976 (25%)0.85440.0042 Other COPD709 (18%)729 (19%)0.55880.0133 Overweight, obesity, and other hyperalimentation345 (9%)361 (9%)0.52760.0144BMI (kg/m2) ≤19.9235 (6%)270 (7%)0.10720.0367 20-29402 (10%)434 (11%)0.24120.0267 30-39158 (4%)153 (4%)0.77230.0066 ≥4027 (0.7%)27 (0.7%)1.00000.0000BMI = body mass index, CKD = chronic kidney disease, COPD = chronic obstructive pulmonary disease, SMD = standardized mean difference Matched Patient Characteristics—Femoral Neck Fracture Repair Cohorts BMI = body mass index, CKD = chronic kidney disease, COPD = chronic obstructive pulmonary disease, SMD = standardized mean difference Bar graph showing femoral neck fracture repair mortality risk. Table 4Femoral Neck Fracture Repair Secondary 90-Day OutcomesNo Dementia (n = 3682)(% Risk)Dementia (n = 3682)(% Risk)RR (95% CI)P Value90-d outcomes Readmission1336 (36%)1610 (43%)0.83 (0.78-0.88) <0.0001 Joint infection26 (0.7%)38 (1%)0.68 (0.42-1.12)0.1300 Pneumonia157 (5%)195 (6%)0.78 (0.63-0.95) 0.0153 Cardiovascular events105 (3%)105 (3%)0.98 (0.75-1.28)0.9080 Stroke80 (2%)104 (3%)0.72 (0.54-0.96) 0.0229 UTI265 (10%)294 (14%)0.71 (0.61-0.83) <0.0001 Revision arthroplasty44 (1%)26 (0.7%)1.69 (1.04-2.74) 0.0308 Nonunion60 (2%)43 (1%)1.40 (0.95-2.06)0.0907 Malunion25 (1%)18 (0.5%)1.39 (0.76-2.54)0.2840 Heart failure179 (6%)139 (5%)1.28 (1.04-1.59) 0.0223 Respiratory complications211 (7%)184 (6%)1.13 (0.93-1.36)0.2240Bold entries are significant p-value.CI = confidence interval, RR = risk ratio, UTI = urinary tract infection Femoral Neck Fracture Repair Secondary 90-Day Outcomes Bold entries are significant p-value. CI = confidence interval, RR = risk ratio, UTI = urinary tract infection Discussion Hip fractures are among the most common orthopaedic injuries in the elderly population and often require timely surgical intervention to prevent disability and mortality.15 This presents a growing challenge for current and future orthopaedic surgeons. As the global population continues to age, there will be a corresponding increase in hip fracture cases requiring treatment. Estimates indicate that by 2050, nearly 58% of countries worldwide will have populations comprising 14% or more individuals aged 65 years and older, compared with 2020, when fewer than 30% of countries had populations of this age.16 Similarly, 300,000 US adults aged 65 years and older fracture a hip annually, mostly from falls.10 Compounding these issues is the rising prevalence of dementia, which has been projected to increase by nearly 50% from the year 2020 to 2060.17 Given that the neurocognitive effects of dementia can markedly impair gait, a likely consequence is an increase in falls resulting in hip fractures that necessitate surgical intervention.18,19 Although surgery has proven effective in managing hip fractures, patients often fail to regain full mobility, leading to recurrent falls and impairments in activities of daily living.20,21 The added comorbid burden of dementia in these cases can contribute to worse outcomes and increased mortality. In the context of this information, this large multicenter retrospective cohort study found that preexisting dementia is markedly associated with an increased postoperative mortality following surgical repair of both intertrochanteric and femoral neck hip fractures in adults aged 65 to 90 years. These findings support our original hypothesis and reinforce prior evidence, suggesting that dementia is an independent predictor of poor postoperative outcomes and increased mortality. Given the complications dementia imposes on outcomes, increasing attention has been directed toward identifying specific causes of morbidity and mortality associated with dementia. For example, a study by Ioannidis et al22 used a similar retrospective design on a single-country cohort and found that patients with dementia undergoing hip fracture repair were at an increased risk of death from postoperative cardiovascular, respiratory, cerebrovascular events, and sepsis. Although their study focused on 30-day mortality, the results align with this study, which also found a notable increase in postoperative mortality among patients with preexisting dementia. Although their study did not specify diagnosis codes for cause-specific mortality, these events were notably more common in individuals with dementia. Similarly, this study observed modestly increased risks of pneumonia and stroke in dementia patients across both fracture types. No difference was seen in cardiovascular events, likely due to differing event definitions and follow-up periods. The longer 90-day follow-up period may have captured more events in both groups, thereby narrowing the RR and ultimately contributing to the lack of statistical significance. In a similar study looking at postoperative mortality in Korea, Jeong et al23 noted that at a 1-year follow-up, patients with dementia were 1.23 times more likely to have a mortality event than those without dementia. Likewise, Bai et al12 in their meta-analysis of 18 studies between 2000 and 2018 published results showing that dementia increased mortality by nearly two-fold at 30 days, 6 months, and 1 year after hip fracture surgery. In a recent study comparing postoperative outcomes in THA and total knee arthroplasty (TKA), Johnson et al24 reported a 1.75 times increased likelihood of readmission in patients with dementia, which our study supports (intertrochanteric fracture: 0.95 times more likely; femoral neck fracture repair: 0.83 times more likely). Interestingly, their study found no difference in revision arthroplasty in either THA or TKA groups, which differs from the results of our study, showing that patients without dementia had an increased risk of undergoing revision arthroplasty within 90 days. Although perioperative management of elective TKA and THA differs from that of acute hip fractures, the impact of dementia on select patient outcomes appears to be similar. Another study examining prosthesis complications found no difference in revisions when focusing specifically on patients after femoral neck fracture surgery.13 It should be noted that the risk difference is small in both intertrochanteric and femoral neck fractures (0.27%, P = 0.0001 and 0.47%, P = 0.0308), indicating that this result may be statistically significant but not clinically meaningful. Ahluwalia et al examined similar outcomes at 90 days in patients undergoing primary THA for femoral neck fractures. They were able to show a markedly higher risk of several outcomes in patients with dementia, including cerebrovascular accidents, pneumonia, and myocardial infarction, with the addition of other outcomes such as acute kidney injury, deep vein thromboses, and venous thromboembolism.13 Finally, in their retrospective study, which used similar ICD 10th edition and CPT codes, Tsuda et al who used a national administrative database in Japan to examine more than 87,000 elderly hip fracture patients. Their analysis showed that dementia independently increases the risk of complications such as pneumonia and UTIs, which are similar to the results found in our multicenter analysis.25 An increased risk of postoperative heart failure was seen in patients without dementia in this study. Several risks have been noted to cause postoperative heart failure, such as age, hypertension, anemia, and duration of surgery.26 These variables have mainly been studied in the acute postoperative setting of hip fracture, whereas our study showed significance out to 90 days. The impact of these risk factors for postoperative heart failure needs to be studied beyond the acute postoperative setting of hip fracture to further assess the exact cause for the increased risk of heart failure in patients without dementia undergoing hip fracture repair. Although this study implemented several mechanisms to minimize bias and errors, it is not without limitations. Retrospective study designs may introduce the potential for selection bias and misclassification, despite using the propensity score matching to balance for baseline characteristics. Because the TriNetX database uses real-world electronic health record (EHR) data, inaccuracies may be introduced in patient coding, which could confound the results of the study. Furthermore, the use of ICD and CPT codes does not always capture the entire clinical picture of a patient or allow for important metrics to be assessed. For example, the codes used to define dementia in this study were unable to specifically track the severity of dementia, leaving a notable aspect of a patient's disease that could alter the results. Similarly, differences between dementia types were not assessed, which may have left out some crucial differences in outcome comparisons. In addition, the database did not include detailed perioperative data such as time to surgery or the type of anesthesia used. Since time to surgery is a known predictor of mortality and complications, patients with dementia might experience longer delays due to the need for medical optimization or the complexities involved in obtaining consent. A final limitation of this study is the exclusion of patients treated with THA. Although this was done to maintain homogeneity in surgical intent and to avoid procedural variability, it may limit the generalizability of our findings to the subset of hip fracture patients who receive THA. Conclusion This large multicenter cohort database study highlights the impact of preexisting dementia on postoperative mortality in patients undergoing surgical repair of intertrochanteric and femoral neck fractures in older adults at 90 days, 2, and 5 years—with mortality reaching double the rate among patients with dementia at 90 days. However, it is possible that the increased risk of mortality at 2 and 5 years could represent the known increased mortality risk for patients with dementia and not be explicitly a result of the hip fracture. Meanwhile, the marked increase in the first 90 days more likely represents the actual increased mortality from the hip fracture. In addition, patients with dementia were also observed to be at an increased risk of readmission, pneumonia, UTI, and stroke at 90 days postoperatively. These findings can be used to better advise patients, caregivers, and care teams on enhancing shared decision making for the optimal care of patients with dementia. Future studies should investigate the severity and specific types of dementia as potential variables in these outcomes. In addition, multidisciplinary models of care can be developed and tested to optimize outcomes.