Context.—

Autopsy rates have decreased dramatically despite providing important clinical information to medical practices and social benefits to decedents' families.

Objective.—

To assess the impact of an institutional Office of Decedent Affairs (ODA), a direct communication link between pathology and decedents' families, on hospital autopsy consent rates, autopsy-related communication, practitioner views, and next-of-kin experiences.

Design.—

A before and after study involving all hospital decedents whose deaths did not fall within the jurisdiction of the medical examiner's office from 2013 to 2018. A pathology-run ODA launched in May 2016 to guide next-of-kin through the hospital death process (including autopsy-related decisions) and serve as the next-of-kin's contact for any subsequent autopsy-related communication. Critical care and hematology/oncology practitioners were assessed for their autopsy-related views and decedents' next-of-kin were assessed for their autopsy-related experiences. Autopsy consent rates for non–medical examiner hospital deaths, autopsy-related communication rates, practitioner views on the role and value of autopsy, and next-of-kin autopsy experiences and decisions factors were compared prior to and after ODA launch.

Results.—

Autopsy consent rates significantly increased from 13.2% to 17.3% (480 of 3647 deaths versus 544 of 3148 deaths; P < .001). There were significant increases in the rate of autopsy-related discussions and bereavement counseling provided to decedents' families. Practitioner views on the positive role of autopsy for any hospital death and those with advanced stage cancer also significantly increased. Next-of-kin indicated more consistent autopsy-related discussions with the potential benefits of autopsy discussed becoming key decision factors.

Conclusions.—

An ODA improves hospital autopsy consent rates, autopsy-related communication, providers' autopsy-related views, and next-of-kins autopsy experiences.

Since the early 1970s, autopsy rates in the United States have steadily declined to below 10%, driven predominantly by the decline of autopsies after disease-based or non–medical examiner (non-ME) deaths.1  In the early 1970s, the Joint Commission on Accreditation of Healthcare Organizations eliminated minimum autopsy rate requirements for teaching and non-teaching hospitals (previously set at 25% and 20%, respectively) allowing autopsy rates to fall for the first time below those levels.2 

The lack of a minimal standard, though, does not itself drive down autopsy rates. Additional factors negatively affecting autopsy rates have been posited, including lack of clinician knowledge about the autopsy process and the value it provides, perceptions of potential legal implications for the treating medical team, and cost.27  Autopsies, though, remain value-added endeavors. Large autopsy series have shown 15% to 40% of autopsies have significant unexpected findings and clinician-filed death certificates agree with post-autopsy pathologist-filed death certificates less than half the time.811  Furthermore, studies have consistently shown a lack of correlation between autopsies and negative legal consequences for the medical team.5,12  Most importantly, though, autopsies can provide comfort to families by answering unresolved questions and collecting tissue for future research.36 

We hypothesize the decline in hospital autopsy rates is driven in part by a lack of direct, effective communication between the pathologist and the decedent's family. To that end, an Office of Decedent Affairs (ODA) directly connects pathology-trained staff with the decedent's next-of-kin to help guide them through the hospital death process, including connecting next-of-kin with organ donation staff and discussing decisions regarding autopsy and funeral arrangements. The ODA can also serve as the next-of-kin's primary contact for all subsequent communications regarding these issues. In the 1980s and 1990s, an ODA was attributed to stabilizing hospital autopsy rates at the University of Texas Medical Branch at Galveston.1316  This study aimed to assess the impact of a pathology-run ODA in today's healthcare system on the following parameters: (1) autopsy consent rates for non-ME hospital deaths, (2) autopsy-related communication, (3) practitioners' autopsy-related attitudes and beliefs, and (4) next-of-kins autopsy-related experiences.

METHODS

Study Design

This study was performed at a 2000+ bed academic tertiary medical facility that serves as the regional ME's office. Prior to ODA launch, treating medical team members were expected to complete standardized paper forms upon a patient's death, connect next-of-kin with organ donation staff, and have conversations with next-of-kin regarding autopsy and funeral arrangements (Figure 1). After an autopsy, final results were conveyed to the treating medical physician of record who was expected to contact the next-of-kin regarding the results. Any incoming autopsy-related communication from families was directed to the treating medical physician of record.

Figure 1

Autopsy-related communication and workflow for hospital deaths before and after Office of Decedent Affairs launch. For activities shared across parties (spanning >1 swim lane), the side from which the arrow originates indicates the responsible party for initiating the next activity. Dashed lines indicate further communication and workflow if autopsy consent is obtained. Abbreviation: ODA, Office of Decedent Affairs.

Figure 1

Autopsy-related communication and workflow for hospital deaths before and after Office of Decedent Affairs launch. For activities shared across parties (spanning >1 swim lane), the side from which the arrow originates indicates the responsible party for initiating the next activity. Dashed lines indicate further communication and workflow if autopsy consent is obtained. Abbreviation: ODA, Office of Decedent Affairs.

The ODA is organizationally embedded in the Department of Laboratory Medicine & Pathology within the work unit of Postmortem Services, which also includes the Autopsy Laboratory/Service. It consists of 11 ODA specialists filling 8.3 full-time equivalents to provide 24 hours per day, 7 days per week, 365 days per year coverage with 2 ODA specialists on service at all times. ODA specialists are required to have at least a 4-year degree in human or social sciences, but most have a social science/mortuary background.

After ODA launch on May 1, 2016, the ODA is contacted upon any hospital death, comes to the decedent's hospital location to meet with next-of-kin, and completes the aforementioned tasks once performed by treating medical team members (Figure 1). The hand-off of significant postmortem paperwork and discussions with family to ODA subject matter experts allows the medical team to more promptly return to their primary clinical duties. The ODA also coordinates final autopsy reports being sent to next-of-kin by autopsy staff and any incoming autopsy-related communication from families is directed to the ODA.

Participants and Measurements

Non-ME Hospital Autopsy Consent Rates

Data on all hospital non-ME deaths from 2013 to 2018 were extracted from the anatomic pathology laboratory information system (Cerner CoPath Plus, Kansas City, Missouri) and each death classified by date of death and autopsy consent status.

Autopsy-Related Process and Communication Rates

To further explore the ODA's impact on autopsy-related processes and communication, a pre-ODA study cohort (all non-ME hospital deaths between June 1, 2013 and May 31, 2014) and post-ODA study cohort (all non-ME hospital deaths between July 1, 2016 and June 30, 2017) were further explored. For each study cohort member, the institutional electronic health record (GE Centricity, Wauwatosa, Wisconsin), paper-based hospital death record, and anatomic pathology laboratory information system (Cerner CoPath Plus) were audited (save for medical records that were access-restricted for security reasons) using a pre-tested abstraction tool developed for the study. Abstracted data included demographics, hospital location, treating medical physician of record, date/time of death, next-of-kin's relationship to the decedent, organ donation eligibility status, anatomic donation status, documented autopsy-related conversations with the decedent and decedent's family, documented organ donation conversations with organ donation staff for eligible decedents, and documented bereavement counseling provided to the decedent's family. Autopsy consenter, unexpected autopsy findings (major being defined as related to cause of death or impacting treatment/management), and any documented post-autopsy communication with the family (including transmission of final autopsy reports to next-of-kin) was abstracted when applicable.

Practitioner Autopsy-Related Attitudes and Beliefs

All attending physicians, clinical fellows, nurse practitioners, and physician assistants in the medical intensive care unit and hematology/oncology service (high mortality clinical areas) were electronically surveyed (, see supplemental digital content at https://meridian.allenpress.com/aplm in the January 2021 table of contents, containing 2 tables). The institution's survey research center administered the surveys. The survey research center sent hyperlinks via email to all potential participants 2 months before (time 1) and 24 months after (time 2) ODA launch. Surveys were available for 4 weeks with weekly reminders from survey research center or study staff. Responses were linked to staff email addresses but only accessible by survey research center staff. Participants received no remuneration for participation.

The survey included questions about demographics, job classification, training/comfort level regarding death/dying-related and autopsy-related discussions, and perceptions regarding the role and value of autopsy ().17  Questions regarding the anticipated/perceived impact of the ODA and those persons believed to be appropriate parties to discuss autopsies were also included. Responses assessed levels of agreement along a 4- or 5-point Likert scale except for questions regarding appropriate persons for discussing autopsies, which were binary agree versus disagree choices.

Next-of-kin Autopsy-Related Experiences

A purposive sampling using a criterion-based strategy identified participants for semistructured interviews. From the pre-ODA and post-ODA cohorts, a representative sample of 177 next-of-kin (pre-ODA = 80; post-ODA = 97) were recruited to participate via letters and emails using contact information from the decedent's hospital death record. Twenty-two participants consented. Interviews were conducted 6 to 8 months after the decedent's death using a semistructured guide developed by the study team. The guide focused on the hospital death process, content/delivery of autopsy-related discussions, and autopsy decision factors (). All interviews were conducted over the telephone and recorded. This study was approved by the local institutional review board with informed consent obtained for the practitioner surveys and next-of-kin interviews.

Statistical Analysis

Non-ME Hospital Autopsy Consent Rates

Hospital non-ME autopsy consent rates were calculated before ODA launch (January 1, 2013 to April 30, 2016) and after ODA launch (May 1, 2016 to December 31, 2018). The overall consent rate difference between these 2 time periods was compared using a X2  test using JMP Pro version 13 (SAS, Cary, North Carolina). A P value < .05 was considered statistically significant.

Autopsy-Related Process and Communication Rates

Pre-ODA and post-ODA study cohort data were collected and managed using REDCap (Research Electronic Data Capture), which is a secure, Web-based application designed to support data capture for research studies, providing (1) an intuitive interface for validated data entry; (2) audit trails for tracking data manipulation and export procedures; (3) automated export procedures for seamless data downloads to common statistical packages; and (4) procedures for importing data from external sources.18  Data were also analyzed using JMP Pro version 13 (SAS). Pre-ODA and post-ODA study cohorts were compared using student's t-test, X2  test, or Fisher exact test. For demographic and autopsy-related communication comparisons, a Bonferroni correction factor of 8 or 10, respectively, was applied to calculated P values. A P value < .05 was considered statistically significant.

Practitioner Autopsy-Related Attitudes and Beliefs

Survey respondent data were analyzed using X2  test or Fisher exact test. Survey responses captured on a 4-point Likert scale were reduced to agree (strongly agree and somewhat agree) or disagree (strongly disagree and somewhat disagree) and responses captured on a 5-point Likert scale were reduced to too little/uncomfortable, too much/comfortable, or about right/neither for analysis. For demographic, communication training/comfort level, and autopsy attitude/beliefs comparisons, a Bonferroni correction factor of 14, 8, or 40, respectively, was applied to calculated P values. A P value < .05 was considered statistically significant. Full study data were available to 3 study investigators (JEJ, JJM, MCA).

Next-of-kin Autopsy-Related Experiences

All interviews were transcribed verbatim, de-identified, and checked for accuracy. Qualitative data were analyzed using a modified constant comparison approach.1921  This process begins with an inductive constant coding process to develop initial categories for data reduction. Interview participants were grouped into 4 cohorts as follows: pre-ODA who consented to autopsy (n = 4), pre-ODA who did not consent to autopsy (n = 6), post-ODA who consented to autopsy (n = 6), and post-ODA who did not consent to autopsy (n = 6). The following 3-stage analysis approach was used for group comparisons: (1) within a single participant's interview to summarize core interview components; (2) responses within the same cohort to further reduce and recode data into core categories; and (3) responses across different cohorts. As clarification was sought on major themes, the interviewer returned to the original transcripts to further assess these themes. Recruitment for each group was ceased once thematic saturation was reached.22,23  Analyses and data management were performed using NVivo100 (QSR International Inc., Burlington, Massachusetts), a qualitative software package that facilitates coding, categorizing, comparing, and reporting data for specific themes of interest.

Role of Funding Source

A grant from the National Institutes of Health (NIH UL1 TR000135) provides use of the REDCap application for any institutional research and quality studies. The National Institutes of Health had no role in the design, conduct, or reporting of this study.

RESULTS

Non-ME Hospital Autopsy Consent Rates

Of 3647 non-ME hospital deaths occurring prior to ODA launch, there was consent to having an autopsy in 480 (13.2%). Of 3148 non-ME hospitals deaths occurring after ODA launch, 544 (17.3%) consented to having an autopsy performed (P < .001; Figure 2). The highest autopsy consent rates occurred during the first 1.5 years after ODA launch.

Figure 2

Quarterly autopsy consent rates for non–medical examiner hospital deaths from 2013 to 2018. Office of Decedent Affairs launch date and time periods for study cohorts are indicated. Abbreviation: ODA, Office of Decedent Affairs.

Figure 2

Quarterly autopsy consent rates for non–medical examiner hospital deaths from 2013 to 2018. Office of Decedent Affairs launch date and time periods for study cohorts are indicated. Abbreviation: ODA, Office of Decedent Affairs.

Autopsy-Related Process and Communication Rates

For 652 non-ME hospital deaths occurring within the pre-ODA study cohort, 101 (15.5%) consented to having an autopsy performed. For 849 non-ME hospital deaths occurring within the post-ODA study cohort, 208 (24.5%) consented to having an autopsy performed. Eight high-security medical records were not abstracted (5 consented to autopsy and 3 did not; 7 pre-ODA and 1 post-ODA). Of abstracted records, there were no significant differences in age, sex, county, hospital location, date/time of week, next-of-kin type, or anatomic donation rates between the pre-ODA and post-ODA cohorts (Table 1). There was a significant difference in documentation of organ donation eligibility rates on hospital death forms with greater rates of unknown organ donation eligibility status in the pre-ODA cohort and greater rates of organ donation ineligibility in the post-ODA cohort. The pre-ODA cohort represented the patients of 241 different treating physicians of record while the post-ODA cohort represented the patients of 297 different treating physicians of record.

Table 1

Study Cohort Demographic Information

Study Cohort Demographic Information
Study Cohort Demographic Information

There were no significant differences in the rates of documented patient autopsy-related discussions, organ donation discussions with organ donation staff for those eligible to donate, or bereavement counseling provided to decedent families by the treating medical team between the pre-ODA and post-ODA cohorts (Table 2). There was a significant increase in the rate of decedent family autopsy-related discussions from 570 of 645 (88%) non-ME hospital deaths in the pre-ODA cohort to 840 of 848 (99%) non-ME hospital deaths in the post-ODA cohort (P < .001). Moreover, the ODA provided bereavement counseling to 833 of 848 (98%) decedent families in the post-ODA cohort compared with treating medical teams providing bereavement counseling to 313 of 645 (49%) decedent families in the pre-ODA cohort and 387 of 848 (46%) decedent families in the post-ODA cohort.

Table 2

Study Cohort Documented Autopsy-Related Communication and Findings

Study Cohort Documented Autopsy-Related Communication and Findings
Study Cohort Documented Autopsy-Related Communication and Findings

Among 207 non-hospital ME deaths in the post-ODA cohort for which there was consent for an autopsy, the ODA performed the autopsy consent in 198 instances (96%; Table 2). There was a significant decrease in the rate that treating medical teams had final autopsy-related communication with the family and a significant increase in the rate that pathology/ODA teams had in the final autopsy-related communication with the family (P ≤ .001). Overall, the rate of any final autopsy-related communication rose from 38 of 97 consented autopsies in the pre-ODA cohort (39%) to 184 of 207 consented autopsies in the post-ODA cohort (89%; P < .001). There was no significant difference in the rate of unexpected autopsy findings or the percentage of autopsy reports completed within 60 days.

Practitioner Autopsy-Related Attitudes and Beliefs

There was a significantly higher response rate for the pre-ODA survey (86 of 162; 53%) than the post-ODA survey (40 of 178 [22%]; P < .001). There was a greater percentage of attending physicians responding to the post-ODA survey than the pre-ODA survey (31 of 83; 37% pre-ODA respondents versus 28 of 37; 75% post-ODA respondents; P = .007; Table 3). Among survey respondents, each individual survey question had a greater than 90% response rate. There was no significant difference in survey respondents with respect to age group, sex, time in practice, percentage of practice devoted to patient care, physician subspecialty, or training/comfort level for death/dying-related and autopsy-related discussions (Table 3; Figure 3).

Table 3

Practitioner Survey Demographics

Practitioner Survey Demographics
Practitioner Survey Demographics
Figure 3

Practitioners' perceived training and comfort levels for death/dying-related and autopsy-related conversations with patients and families before and after Office of Decedent Affairs (ODA) launch. No significant differences were observed between the pre-ODA and post-ODA cohorts after adjustment for multiple comparisons.

Figure 3

Practitioners' perceived training and comfort levels for death/dying-related and autopsy-related conversations with patients and families before and after Office of Decedent Affairs (ODA) launch. No significant differences were observed between the pre-ODA and post-ODA cohorts after adjustment for multiple comparisons.

The majority of pre-ODA respondents agreed that autopsy had a role in prenatal, pediatric, intensive care, early stage cancer, and procedure-related hospital deaths (Figure 4). The majority of post-ODA respondents held similar beliefs but showed significant increases in the percentages of respondents who agreed that autopsy had a role in all hospital deaths (P = .004) and the deaths of advanced stage cancer patients (P = .048).

Figure 4

Practitioner survey responses on the role, value, and common concerns regarding autopsy, anticipated/perceived impact of the Office of Decedent Affairs (ODA), and the appropriate parties to discuss autopsy. *Statistically significant difference between pre-ODA and post-ODA survey respondents after adjustment for multiple comparisons.

Figure 4

Practitioner survey responses on the role, value, and common concerns regarding autopsy, anticipated/perceived impact of the Office of Decedent Affairs (ODA), and the appropriate parties to discuss autopsy. *Statistically significant difference between pre-ODA and post-ODA survey respondents after adjustment for multiple comparisons.

Overall, the majority of pre-ODA and post-ODA survey respondents agreed that autopsy had value for a wide range of purposes (Figure 4). Moreover, the majority of survey respondents would consent to their own autopsy or the autopsy of a family member. Most survey respondents also did not agree with common concerns regarding autopsy.17  There was no significant difference between the anticipated and perceived impact of the ODA and no significant differences in the types of persons deemed appropriate to discuss autopsy with the decedent's family. Despite the ODA being pathology-run, only 10 of 40 (25%) post-ODA practitioners believed that pathology staff were appropriate persons to discuss autopsy.

Next-of-Kin Autopsy-Related Experiences

Among 22 next-of-kin interviewed, the average decedent age was 67.5 years (range, 13–95 years) and 17 (77%) of the decedents were male. Seventeen next-of-kin were the decedent's spouse, 2 next-of-kin were decedent's parent, and the remaining 3 next-of-kin were the decedent's fiancé, son, and niece. The average length of next-of-kin interview was 20 minutes (range, 17–58 minutes).

When asked about the content of autopsy-related conversations, pre-ODA next-of-kin indicated, at most, minimal limited discussions about performing an autopsy (Tables 4 and 5). For those next-of-kin who consented, next-of-kin initiated those conversations due to their prior experiences or knowledge of autopsy. Post-ODA next-of-kin reported thorough, detailed autopsy discussions regardless of whether next-of-kin ultimately consented to autopsy or not.

Table 4

Next-of-Kin Interview Summaries About Autopsy-Related Discussions by Autopsy Consent and ODA Status

Next-of-Kin Interview Summaries About Autopsy-Related Discussions by Autopsy Consent and ODA Status
Next-of-Kin Interview Summaries About Autopsy-Related Discussions by Autopsy Consent and ODA Status
Table 5

Representative Next-of-Kin Statements

Representative Next-of-Kin Statements
Representative Next-of-Kin Statements

Pre-ODA next-of-kin emphasized the importance of having more than 1 family member present and providing privacy during these discussions (Tables 4 and 5). Moreover, showing empathy and presenting autopsy in a noncoercive manner was critical. For post-ODA next-of-kin, delivery of autopsy-related conversations focused instead on the use of multiple information modalities while emphasizing that optimal timing for such conversations varies greatly.

Altruism was a key decision factor in consenting to autopsy both pre-ODA and post-ODA (Tables 4 and 5). Post-ODA next-of-kin who consented to autopsy also indicated the potential benefits of autopsy (including possible genetic implications for other family members) were influential. For those pre-ODA next-of-kin who declined autopsy, the lack of other family members in the decision-making process was an important decision factor. For those post-ODA next-of-kin who declined autopsy, sufficient pre-existing information about the decedent's cause of death and concerns about perceived autopsy costs were cited as important decision factors.

DISCUSSION

Overall, the launch of an ODA led to a 30% relative increase in autopsy consent rates for non-ME hospital deaths, with the largest increases occurring during the first 18 months after ODA launch (Figure 2). In-depth analysis of pre-ODA and post-ODA study cohorts revealed this change occurred with a similar decedent population being treated by an even greater number of practitioners. Practitioner surveys showed both satisfaction with and approval of an ODA. Both the autopsy consent rate increase and practitioner's overall positive responses to the ODA are consistent with the success of early ODAs in the 1980s (Figures 2 and 4).15,16  Increases in non-ME hospital autopsy rates did not alter the rate of unexpected autopsy findings or the percentage of final autopsy reports completed within 60 days (benchmark used by the National Association of Medical Examiners). Despite the ODA's apparent success, practitioners' views regarding pathology staff as appropriate persons to discuss autopsy did not change, suggesting practitioners may not associate the ODA with pathology.

The hypothesis that the lack of direct, effective autopsy-related communication between the pathologist and the decedent's family adversely impacts consent rates is supported by the analysis of autopsy-related communication and next-of-kin experiences. When the autopsy consent process lies with hundreds of different treating practitioners with potentially varying levels of autopsy knowledge, the process can be highly inconsistent. Before the ODA, there were no documented autopsy-related conversations with decedents' families 12% of the time and there was no documented bereavement counseling provided to families 51% of the time. Next-of-kin recollected brief, closed-ended conversations regarding autopsy and the impetus was largely on next-of-kin to actively pursue autopsy. With ODA implementation, documented autopsy-related conversations with decedents' families rose significantly. While documented bereavement counseling rates by the treating medial team did not significantly change, the ODA provided documented bereavement counseling to 98% of families. Regardless of their decision to pursue autopsy or not, next-of-kin recalled thorough autopsy-related discussions, including its potential benefits. In fact, those same benefits appeared as key decision factors for post-ODA next-of-kin consenting to autopsy. While pre-ODA next-of-kin had concerns about having private, empathetic, and noncoercive autopsy-related discussions, those concerns did not persist post-ODA.

As pre-ODA autopsy-related conversations were carried out by treating medical team members, their beliefs about autopsy may have influenced if and how such conversations occurred. Next-of-kin indicated the recommendations of the treating medical team were influential decision factors for autopsy consent. Yet pre-ODA, the majority of practitioners disagreed that autopsy had a role in all hospital deaths, advanced stage cancer deaths of patients, and hospice deaths despite robust consensus around the many values of autopsy. Significant increases in practitioners' agreement on the role of autopsy in all hospital deaths and advanced stage cancer deaths post-ODA suggest the ODA's impact on the hospital autopsy process included practitioners' views about autopsy.

The ODA also significantly improved post-autopsy communication. Similar to the pre-ODA autopsy consent process, final autopsy-related communication with families can be inconsistent. The assumption of these final steps by ODA and pathology staff led to higher overall final autopsy-related communication rates with a corresponding decrease in final autopsy-related communication by treating medical teams.

The ODA has also been equally involved in guiding decedent's families through the autopsy process for those decedents whose hospital deaths legally fell within the jurisdiction of the ME's office. In these instances, ODA staff serve not only as a communication link between the decedent's family and pathology but also the medical examiner's office. Moreover, ODA staff pursue autopsy consent in these cases to facilitate additional medical education and research opportunities. When these additional consented medical examiner hospital cases are included, the ODA's monthly autopsy consent rate runs 2% to 4% higher than shown in Figure 2.

An unexpected finding from this work is that the increase in non-ME hospital autopsy consent rates waned after 18 months, which coincided with the first ODA staff turnovers. All initial ODA staff received 3 months of dedicated training prior to ODA launch. In contrast, subsequently hired ODA staff was given shorter, less intensive training programs. In addition, temporary leaves of absence of other ODA staff during 2018 required temporary staff be hired to fill scheduling gaps during nights and weekends, which are particularly challenging times to have direct contact with families. Autopsy consent rates among these new and temporary ODA staff have been lower and are currently the focus of ongoing quality improvement work.

This study has several limitations. First, the study was conducted at a single-academic medical center that initially had an above average non-ME hospital autopsy consent rate. Second, the hiring and training of sufficient personnel to staff an institution-wide, 24-hour, 7 days per week service can be challenging for large medical institutions and infeasible for smaller organizations, which may limit the generalizability of these findings. Third, the time period used to define the pre-ODA study cohort was prior to any significant institutional development of the ODA. This was done to minimize any potential confounding effects, but resulted in a 2-year time period between study cohorts. While the overall hospital non-ME autopsy consent rate remained stable during this interim period, other unrealized changes may have occurred. Fourth, assessment of autopsy-related communication relied on medical documentation and, as such, may not have captured all such occurrences both pre-ODA and post-ODA. Fifth, the post-ODA practitioner survey had a significantly lower response rate than the pre-ODA survey, but had a higher response rate among attending physicians, which may have influenced survey results. Finally, next-of-kin interviews relied on recollections around a family member's death, which are susceptible to recall bias. The use of a structured interview following similar lengths of time after the decedent's death was used to mitigate any recall bias.

CONCLUSIONS

In summary, an ODA produced a significant increase in non-ME hospital autopsy consent rates along with significant increases in the rate of documented autopsy-related discussions with families and the rate at which next-of-kin received final autopsy-related communication. Practitioners' positive views on the role of autopsy for all hospital deaths and advanced stage cancer deaths showed significant increases. Next-of-kin indicated more consistent, thorough discussions about the potential benefits of autopsy and those potential benefits translated into key decision factors for those consenting to autopsy.

The authors would like to thank our institutional Survey Research Center for its assistance in executing this study and the National Institutes of Health (NIH UL1 TR000135) for financial support for the use of the REDCap application in this study. Finally, Dr Griffin's effort was supported by the Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery.

References

References
1.
Hoyert
DL.
The changing profile of autopsied deaths in the United States, 1972-2007
.
NCHS Data Brief
.
2011
;
(67)
:
1
8
.
2.
Rosenbaum
GE,
Burns
J,
Johnson
J,
Mitchell
C,
Robinson
M,
Truog
RD.
Autopsy consent practice at US teaching hospitals: results of a national survey
.
Arch Intern Med
.
2000
;
160
(3)
:
374
380
.
3.
Oppewal
F,
Meyboom-de Jong
B.
Family members' experiences of autopsy
.
Fam Pract
.
2001
;
18
(3)
:
304
308
.
4.
Hull
MJ,
Nazarian
RM,
Wheeler
AE,
Black-Schaffer
WS,
Mark
EJ.
Resident physician opinions on autopsy importance and procurement
.
Hum Pathol
.
2007
;
38
(2)
:
342
350
.
5.
Horowitz
RE,
Naritoku
WY.
The autopsy as a performance measure and teaching tool
.
Hum Pathol
.
2007
;
38
(5)
:
688
695
.
6.
Spunt
SL,
Vargas
SO,
Coffin
CM,
et al.
The clinical, research, and social value of autopsy after any cancer death: a perspective from the Children's Oncology Group Soft Tissue Sarcoma Committee
.
Cancer
.
2012
;
118
(12)
:
3002
3009
.
7.
Burton
JL,
Underwood
J.
Clinical, educational, and epidemiological value of autopsy
.
Lancet
.
2007
;
369
(9571)
:
1471
1480
.
8.
Zarbo
RJ,
Baker
PB,
Howanitz
PJ.
The autopsy as a performance measurement tool–diagnostic discrepancies and unresolved clinical questions: a College of American Pathologists Q-Probes study of 2479 autopsies from 248 institutions
.
Arch Pathol Lab Med
.
1999
;
123
(3)
:
191
198
.
9.
Tavora
F,
Crowder
CD,
Sun
CC,
Burke
AP.
Discrepancies between clinical and autopsy diagnoses: a comparison of university, community, and private autopsy practices
.
Am J Clin Pathol
.
2008
;
129
(1)
:
102
109
.
10.
Tejerina
E,
Esteban
A,
Fernandez-Segoviano
P,
et al.
Clinical diagnoses and autopsy findings: discrepancies in critically ill patients*
.
Crit Care Med
.
2012
;
40
(3)
:
842
846
.
11.
Sington
JD,
Cottrell
BJ.
Analysis of the sensitivity of death certificates in 440 hospital deaths: a comparison with necropsy findings
.
J Clin Pathol
.
2002
;
55
(7)
:
499
502
.
12.
Bove
KE,
Iery
C.
The role of the autopsy in medical malpractice cases, I: a review of 99 appeals court decisions
.
Arch Pathol Lab Med
.
2002
;
126
(9)
:
1023
1031
.
13.
Carter
JR,
Martin
DL.
A pathology assistant program: the role of licensed morticians
.
Am J Clin Pathol
.
1970
;
53
(1)
:
26
31
.
14.
Smith
JH,
Haque
AK,
Gilmer
PR
Jr,
Jekielek
S.
A solution to declining autopsy rates–the Decedent Affairs Office
.
Pathologist
.
1986
;
40
(7)
:
22
24
.
15.
Haque
AK,
Cowan
WT,
Smith
JH.
The decedent affairs office. A unique centralized service
.
JAMA
.
1991
;
266
(10)
:
1397
1399
.
16.
Haque
AK,
Patterson
RC,
Grafe
MR.
High autopsy rates at a university medical center
.
What has gone right? Arch Pathol Lab Med
.
1996
;
120
(8)
:
727
732
.
17.
Oluwasola
OA,
Fawole
OI,
Otegbayo
AJ,
Ogun
GO,
Adebamowo
CA,
Bamigboye
AE.
The autopsy: knowledge, attitude, and perceptions of doctors and relatives of the deceased
.
Arch Pathol Lab Med
.
2009
;
133
(1)
:
78
82
.
18.
Harris
PA,
Taylor
R,
Thielke
R,
Payne
J,
Gonzalez
N,
Conde
JG.
Research electronic data capture (REDCap) –a metadata-driven methodology and workflow process for providing translational research informatics support
.
J Biomed Inform
.
2009
;
42
(2)
:
377
381
.
19.
Boeije
H.
A purposeful approach to the constant comparative method in the analysis of qualitative interviews
.
Qual Quant
.
2002
;
36
(4)
:
391
409
.
20.
Fram
SM.
The constant comparative analysis method outside of grounded theory
.
The Qualitative Report
.
2013
;
18
(1)
:
1025
.
21.
Heydon
R,
Hibbert
K.
“Relocating the Personal” to engender critically reflective practice in preservice literacy teachers
.
Teach Teach Educ
.
2010
;
26
(4)
:
796
804
.
22.
Marshall
MN.
Sampling for qualitative research
.
Fam Pract
.
1996
;
13
(6)
:
522
525
.
23.
Malterud
K,
Siersma
VD,
Guassora
AD.
Sample size in qualitative interview studies: guided by information power
.
Qual Health Res
.
2016
;
26
(13)
:
1753
1760
.

Author notes

This project has been funded with federal funds from an Institutional National Institutes of Health Grant to subsidize enterprise use of the REDCap database tool (NIH UL1 TR000135).

Supplemental digital content is available for this article at https://meridian.allenpress.com/aplm in the January 2021 table of contents.

The authors have no relevant financial interest in the products or companies described in this article.

This work has been presented in abstract form at the United States and Canadian Academy of Pathology annual meetings on March 15, 2016 in Seattle, Washington and March 19, 2018 in Vancouver, British Columbia.

Supplementary data