Abstract
This study conveys data on the prevalence of advanced care planning documentation in elderly individuals undergoing high-risk surgery.
More than 4 million high-risk operations (those that are associated with ≥1% in-hospital mortality) are performed annually in the United States in patients 65 years or older.1 While operative risk has declined over time, many older adults, especially those with multiple chronic conditions, remain at high risk for postoperative morbidity (including loss of independence and/or functional decline) and mortality.2,3 Therefore, the American College of Surgeons and American Geriatric Society jointly recommend patients engage in advance care planning (ACP), which includes documenting a patient’s personal goals and values, treatment preferences, and surrogate decision maker.4 In this study, we determined the prevalence and patient characteristics associated with ACP documentation among elderly adults at any time prior to high-risk surgery.
Methods
We included patients 65 years and older who were evaluated at Sutter Health (Palo Alto Medical Foundation), from January 1, 2013, through December 31, 2014. Eligible individuals were those who had multiple chronic conditions (defined as a Charlson Comorbidity score >15) and who underwent a high-risk procedure (defined by validated surgical Current Procedural Terminology codes that are associated with an in-hospital mortality rate ≥1%1). This study was approved by the Sutter Health institutional review board. Demographic and clinical data were abstracted from electronic health records; no informed consent was needed.
The primary outcome of ACP documentation was abstracted from health records, because Palo Alto Medical Foundation collects this information in an electronic problem list. Palo Alto Medical Foundation uses an automated clinician reminder that is triggered when patients turn 65 years old and continues to fire at all appointments until completed. Documentation of ACP included the presence of either an advanced directive, a durable power of attorney for health care, or a physician order for life-sustaining treatment. In this study, no distinction was made as to the timing of ACP documentation as long as it was completed prior to the index surgery.
We conducted descriptive and comparative statistics (χ2 test for categorical variables) with Stata version 14.2 (StataCorp), considering P values less than .05 statistically significant. We assessed unadjusted and adjusted probability of ACP documentation across specific patient characteristics and used logistic regression to simultaneously adjust for a parsimonious selection of associated variables. We also reviewed 25 randomly selected electronic health records of patients who died within a year of surgery to assess whether any preoperative notes referenced the patient’s personal goals and values in developing the surgical care plan. We followed up all included patients from study inception to their death or December 31, 2014, whichever came first. Data analysis was conducted from November 2016 through October 2017.
Results
Overall, 393 patients (mean [SD] age, 79.0 [7.8] years) met inclusion criteria (Table 1). Of the overall cohort, 101 (25.7%) had ACPs documented preoperatively. Among those who died within a year of surgery (n = 55 [14.0%]), only 17 (30.9%) had documentation. Mean (SD) time to death after surgery was 107 (104) days (range, 4-341 days).
Table 1. Demographics and Baseline Characteristics of Patients by Advance Care Planning Documentationa.
| Characteristic | No. (%) | |||
|---|---|---|---|---|
| Total (N = 393) | Had ACP Documented Prior to Index Surgery (n = 101) | No ACP Documented Prior to Index Surgery (n = 292) | P Value | |
| Age group, y | ||||
| 65-74 | 125 (31.8) | 24 (19.2) | 101 (80.8) | .08 |
| 75-84 | 166 (42.2) | 44 (26.5) | 122 (73.5) | |
| ≥85 | 102 (26.0) | 33 (32.4) | 69 (67.6) | |
| Sex | ||||
| Male | 216 (55.0) | 56 (25.9) | 160 (74.1) | .91 |
| Female | 177 (45.0) | 45 (25.4) | 132 (74.6) | |
| Language | ||||
| English | 360 (91.6) | 95 (26.4) | 265 (73.6) | .30 |
| Non-English | 33 (8.4) | 6 (18.2) | 27 (81.8) | |
| Marital status | ||||
| Married/partnered | 232 (59.0) | 61 (26.3) | 171 (73.7) | .34 |
| Single | 115 (29.3) | 25 (21.7) | 90 (78.3) | |
| Other/unknown | 46 (11.7) | 15 (32.6) | 31 (67.4) | |
| Office visits in y prior to surgery, No. | ||||
| 0-3 | 89 (22.6) | 4 (4.5) | 85 (95.5) | <.001 |
| 4-7 | 55 (14.0) | 7 (12.7) | 48 (87.3) | |
| ≥8 | 249 (63.4) | 90 (36.1) | 159 (63.9) | |
| Race/ethnicity | ||||
| White | 291 (74.0) | 82 (28.2) | 209 (71.8) | .06 |
| Nonwhite | 102 (26.0) | 19 (18.6) | 83 (81.4) | |
| Insurance type | ||||
| HMO | 76 (19.3) | 21 (27.6) | 55 (72.4) | .66 |
| PPO or FFS | 315 (80.2) | 79 (25.1) | 236 (74.9) | |
| Other/unknown | 2 (0.5) | 1 (50.0) | 1 (50.0) | |
| Dementia or mild cognitive impairmentb | ||||
| Yes | 11 (2.8) | 7 (63.6) | 4 (36.4) | .003 |
| No | 382 (97.2) | 94 (24.6) | 288 (75.4) | |
| Presence of serious illnessc | ||||
| Yes | 121 (30.8) | 40 (33.1) | 81 (66.9) | .03 |
| No | 272 (69.2) | 61 (22.4) | 211 (77.6) | |
| Serious illness types | ||||
| Cancer | 28 (7.1) | 11 (39.3) | 17 (60.7) | .09 |
| Neurologic | 30 (7.6) | 12 (40.0) | 18 (60.0) | .06 |
| Cardiac | 56 (14.2) | 16 (28.6) | 40 (71.4) | .60 |
| Surgical mortality risk, % | ||||
| 1-4.9 | 126 (32.1) | 33 (26.2) | 93 (73.8) | .88 |
| ≥5 | 267 (67.9) | 68 (25.5) | 199 (74.5) | |
| Surgery typed | ||||
| Elective | 153 (38.9) | 31 (20.3) | 122 (79.7) | .05 |
| Emergency | 240 (61.1) | 70 (29.2) | 170 (70.8) | |
| Surgery type | ||||
| Cardiothoracic | 212 (53.9) | 52 (24.5) | 160 (75.5) | .80 |
| Gastrointestinal | 92 (23.4) | 25 (27.2) | 67 (72.8) | |
| Vascular | 61 (15.5) | 18 (29.5) | 43 (70.5) | |
| Othere | 28 (7.1) | 6 (21.4) | 22 (78.6) | |
| Charlson Comorbidity Score | ||||
| 2 | 136 (34.6) | 25 (18.4) | 111 (81.6) | .02 |
| ≥3 | 257 (65.4) | 76 (29.6) | 181 (70.4) | |
| Death within 1 y of surgery | ||||
| Yes | 55 (14.0) | 17 (30.9) | 38 (69.1) | .34 |
| No | 338 (86.0) | 84 (24.9) | 254 (75.2) | |
Abbreviations: ACP, advance care planning; FFS, fee for service: HMO, health management organization; PPO; preferred provider organization.
Baseline characteristics of older adults with multiple chronic conditions who underwent high-risk surgery in the Palo Alto Medical Foundation Health System.
Dementia or cognitive impairment diagnosis was identified based on the International Classification of Diseases, Ninth Revision codes listed in the National Committee for Quality Assurance palliative and end-of-life care measure set for a dementia diagnosis (ie, 290.0, 290.1, 290.2, 290.3, 290.4, 290.8, 290.9, 331.0, 331.1, 331.11, 331.82, and 331.2) and mild cognitive impairment diagnoses (331.83 and 310.8).
Serious illness includes cancer (brain, esophageal, lung, pancreatic, peritoneal, and stomach cancers and secondary malignant neoplasm of respiratory and digestive systems or other specified sites), respiratory disease (lung dysfunction, chronic obstructive pulmonary disease, and cystic fibrosis), cardiac dysfunction (heart failure), and neurologic disorder (amyotrophic lateral sclerosis, dementia, Parkinson disease, and cerebrovascular accident). General palliative care was excluded.
Elective surgery is defined by the presence of a surgery office visit documented in the 6 months prior to the operation. If none was documented in this time frame, the operation was categorized as an emergency.
Other includes surgery procedures such as brain surgery, urological surgery, and hernia repair.
In an adjusted analysis (Table 2), individuals 85 years or older were more likely to have ACP documentation than individuals aged 65 to 74 years (adjusted odds ratio [aOR], 2.00 [95% CI, 1.03-3.90]; P = .04). Similarly, patients with higher health care use (more than 8 office visits in the year before surgery) were more likely to have ACP documentation than those with lower use (0-3 such visits; aOR, 13.01 [95% CI, 4.07-41.64]; P < .001) and those with cognitive impairment, defined as a dementia or mild cognitive impairment diagnosis at the time of surgery were more likely to have ACP documentation than those without cognitive impairment (aOR, 6.12 [95% CI, 1.54-24.31]; P = .01).
Table 2. Multivariate Model of Patient Factors Associated With Advance Care Planning Documentation Prior to High-Risk Surgery.
| Characteristic | Unadjusted | Adjusteda | ||
|---|---|---|---|---|
| Odds Ratios (95% CI) | P Value | Odds Ratios (95% CI) | P Value | |
| Age at date of surgery, y | ||||
| 65-74 | 1 [Reference] | NA | 1 [Reference] | NA |
| 75-84 | 1.52 (0.86-2.67) | .15 | 1.55 (0.84-2.84) | .16 |
| ≥85 | 2.01 (1.09-3.70) | .02 | 2.00 (1.03-3.90) | .04 |
| Marital status | ||||
| Married/partnered | 1 [Reference] | NA | 1 [Reference] | NA |
| Singleb | 0.77 (0.46-1.33) | .36 | 0.77 (0.42-1.40) | .40 |
| Other/unknownc | 1.36 (0.69-2.69) | .38 | 1.49 (0.66-3.34) | .34 |
| Office visits in y prior to surgery, No. | ||||
| 0-3 | 1 [Reference] | NA | 1 [Reference] | NA |
| 4-7 | 3.10 (0.86-11.15) | .08 | 3.75 (0.94-15.03) | .06 |
| ≥8 | 12.03 (4.27-33.92) | <.001 | 13.01 (4.07-41.64) | <.001 |
| Race/ethnicity | ||||
| Nonwhite | 1 [Reference] | NA | 1 [Reference] | NA |
| White | 1.75 (0.98-3.00) | .06 | 1.80 (0.97-3.34) | .06 |
| Dementia or mild cognitive impairment diagnosis at time of surgery | ||||
| No | 1 [Reference] | NA | 1 [Reference] | NA |
| Yes | 5.36 (1.53-18.75) | .01 | 6.12 (1.54-24.31) | .01 |
| Serious illness at time of surgeryd | ||||
| No | 1 [Reference] | NA | 1 [Reference] | NA |
| Yes | 1.71 (1.06-2.75) | .03 | 1.64 (0.96-2.80) | .07 |
| Type of surgerye | ||||
| Emergency | 1 [Reference] | NA | 1 [Reference] | NA |
| Elective | 0.6 (0.37-0.97) | .04 | 0.86 (0.50-1.48) | .58 |
Abbreviation: NA, not applicable.
The variables mortality risk, surgery categories, serious illness types, lung dysfunction, neurologic, cardiac, and end-stage renal disease were excluded from multivariate modeling because they were not statistically significant on bivariate analysis.
Single includes single, widowed, divorced, or separated individuals.
Other/unknown indicates that the insurance type was unknown, Medicaid, or MediCal.
Serious illness includes cancer (brain, esophageal, lung, pancreatic, peritoneal, and stomach cancers and secondary malignant neoplasm of respiratory and digestive systems or other specified sites), respiratory disease (lung dysfunction, chronic pulmonary obstructive disease, and cystic fibrosis), cardiac dysfunction (heart failure), and neurologic disorder (amyotrophic lateral sclerosis, dementia, Parkinson disease, and cerebrovascular accident). General palliative care was excluded.
Elective surgery is defined by the presence of a surgery office visit documented in the 6 months prior to the operation; if none was documented within this time frame, the operation was categorized as an emergency.
On review of the 25 randomly selected records of patients who died within a year, 16 (64%) had preoperative notes. No ACP documentation was found in any of these records.
Discussion
Among a cohort of 393 older adults with multiple chronic conditions who are undergoing high-risk surgery, 101 (25.7%) had preoperative ACP documentation, including only 17 of 55 decedents (30.9%). Yet in a prior study,6 52% of surgeons self-reported having had preoperative ACP discussions. High-risk populations in this study, including patients 85 years and older, those with dementia, and those with greater health care use, were more likely to have ACP documentation. However, all older adults with multiple chronic conditions undergoing high-risk surgery would benefit from having ACP completed with documentation in medical records. Future studies should address the timing of ACP to the surgical visit and the question of whether ACP affects surgical decision making or outcomes. Outreach, education, and system workflow changes to increase ACP engagement in elderly populations are essential to preparing patients and their families prior to potentially life-changing events.
References
- 1.Schwarze ML, Barnato AE, Rathouz PJ, et al. . Development of a list of high-risk operations for patients 65 years and older. JAMA Surg. 2015;150(4):325-331. doi: 10.1001/jamasurg.2014.1819 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Robinson TN, Eiseman B, Wallace JI, et al. . Redefining geriatric preoperative assessment using frailty, disability and co-morbidity. Ann Surg. 2009;250(3):449-455. [DOI] [PubMed] [Google Scholar]
- 3.Lawrence VA, Hazuda HP, Cornell JE, et al. . Functional independence after major abdominal surgery in the elderly. J Am Coll Surg. 2004;199(5):762-772. doi: 10.1016/j.jamcollsurg.2004.05.280 [DOI] [PubMed] [Google Scholar]
- 4.Mohanty S, Rosenthal RA, Russell MM, Neuman MD, Ko CY, Esnaola NF. Optimal perioperative management of the geriatric patient: a best practices guideline from the American College of Surgeons NSQIP and the American Geriatrics Society. J Am Coll Surg. 2016;222(5):930-947. doi: 10.1016/j.jamcollsurg.2015.12.026 [DOI] [PubMed] [Google Scholar]
- 5.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. doi: 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 6.Redmann AJ, Brasel KJ, Alexander CG, Schwarze ML. Use of advance directives for high-risk operations: a national survey of surgeons. Ann Surg. 2012;255(3):418-423. doi: 10.1097/SLA.0b013e31823b6782 [DOI] [PubMed] [Google Scholar]