Download Ethical Considerations in Electronic Surveillance Research for Nursing Homes and more Essays (university) Medical ethics in PDF only on Docsity!
race & ethnicity • fall 2006 611
N
early 2.5 million Americans currently reside in nursing homes and assisted living facilities in the United States, accounting for approxi- mately five percent of persons sixty-five and older.^1 The aging of the “Baby Boomer” generation is expected to lead to an exponential growth in the need for some form of long-term care (LTC) for this segment of the population within the next twenty-five years. In light of these sobering demographic shifts, there is an urgency to address the profound concerns that exist about the quality-of-care (QoC) and quality-of-life (QoL) of this frailest segment of our population. The United States General Accounting Office (GAO) recently reported that,
One in five nursing homes nationwide (about 3, homes) had serious deficiencies that caused resi- dents actual harm or placed them in immediate jeopardy…Moreover, GAO found significant under- statement of care problems that should have been classified as actual harm or higher – serious avoid- able pressure sores, severe weight loss, and multiple falls resulting in broken noses and other injuries…^2
The GAO attributes the underreporting of such prob- lems to: (a) lack of clarity regarding the definition of harm, (b) inadequate state supervisory review of sur- veys, (c) delays in timely investigation of complaints,
and (d) predictability of the timing of annual nursing home surveys. Equally importantly, without methods to continuously record, monitor and document the care of these residents, it is exceedingly difficult to verify resident-specific data reported by nursing homes and review complaint investigations. These tasks would be greatly aided by tools that enable accurate assessments of patient care and treatment. Concerns about abuse and neglect of nursing home residents have led to the introduction of “granny cam” legislation in at least eleven states. 3 Passage of the legis- lation, in any of its several permutations, would essen- tially permit families and legal representatives of nurs- ing home residents to install video cameras in their loved one’s room to monitor their care. Proponents of “granny cams” have argued passionately that the scope and severity of the abuse and neglect problems war- rant such intervention. Their opponents argue that such measures would unduly burden the LTC industry with unmanageable liability insurance costs, demor- alize the already small and dwindling overburdened nursing home staff, and would not necessarily guaran- tee improved care any more than surveillance cameras at convenience markets have succeeded in curtailing crime.^4 The ethical implications of surveillance tech- nologies on the privacy and dignity of LTC residents, personnel and the care processes involved merit more systematic investigation.
Ethical Considerations in the Conduct
of Electronic Surveillance Research
Ashok J. Bharucha, Alex John London, David Barnard,
Howard Wactlar, Mary Amanda Dew, and
Charles F. Reynolds III
Ashok J. Bharucha, M.D., and Charles F. Reynolds, M.D., are from the Advanced Center in Interventions and Services Research for Late-Life Mood Disorders, University of Pittsburgh School of Medicine. David Barnard, Ph.D., is a Professor at the Center for Bioethics and Health Law, University of Pittsburgh. Alex John London, Ph.D., is Associate Professor of Phi- losophy at Carnegie Mellon University. Howard Wactlar, M.S., is from the School of Computer Science, Carnegie Mellon University. Mary Amanda Dew, Ph.D., is Professor of Psychiatry, Advanced Center in Interventions and Services Research for Late-Life Mood Disorders, University of Pittsburgh School of Medicine, and the Center for Bioethics and Health Law, University of Pittsburgh.
612 journal of law, medicine & ethics
INDEPENDENT
In contrast to “granny cams” which are installed in the nursing home resident’s room, the authors are at- tempting to record and analyze the lives of nursing home residents continuously in real-time in the shared spaces of community-based representative demen- tia units (hallways, activity and dining rooms) using video, audio and sensor technologies. CareMedia: Automated Video and Sensor Analysis for Geriatric Care (NSF 0205219) is a project that uses automated speech, image, and natural language processing to cre- ate a rich, indexed, searchable multimedia database in which salient episodes of resident care and functioning are captured and stored for future clinical evaluation (Figure 1).
Figure 1 Conceptual Overview of the
CareMedia Project
Beginning with raw video, audio and sensor data cap- tured via digital cameras, microphones, and sensors, human coders specify events of interest, such as falls,
punching or kicking. Based on the human coding, computer algorithms are developed to automatically detect the event and its contextual characteristics. In this manner, the computer is “trained” to detect sa- lient events, eventually reducing the need for human data coding to only those circumstances that are phe- nomenologically and semantically challenging even for human eyes. Furthermore, irrelevant and redun- dant data can be reduced using techniques developed over the past decade that permit the analysis of large volumes of raw data that previously would have been prohibitive. Thus, this technology has the potential to record, analyze and document QoC and QoL in nursing home settings by:
- Augmenting qualitative observations with quanti- tative dimensions, transforming largely subjective assessments into more measured, objective ones.
- Enriching discrete, human observations with a machine-captured, continuous longitudinal re-
614 journal of law, medicine & ethics
INDEPENDENT
to illustrate a pragmatic approach to resolving ethical dilemmas that arose during the implementation phase of this project.
Privacy: The Legal Perspective Warren and Brandeis, in their seminal writings on the concept of privacy law, have defined privacy as the “right to be let alone.” 7 Subsequently, the legal concept of privacy has differentiated to encompass three major categories: physical privacy, informational privacy, and decisional privacy. 8 The CareMedia project potentially encroaches upon physical and informational privacy by capturing images, voices and activities of nursing home residents and others in shared spaces, but does not appear to impinge upon decisional privacy. The United States Supreme Court has held that the Fourth Amendment (and in some cases, the First, Fifth, Ninth and Fourteenth Amendments) upholds the expecta- tion of privacy in those situations which society would consider reasonable.^9 The question becomes one of ex- plicating the ethical and philosophical underpinnings of reasonableness. Translating the above argument in terms of the LTC setting, what are reasonable “zones of privacy” within such residential facilities, and to what extent do they apply to the various stakeholders identified in Table 1? In the absence of explicit statutes governing rea- sonable zones of privacy within LTC institutions, we are compelled to rely upon court opinions offered in other similar settings for guidance. For example, courts have previously ruled the common areas of residential buildings (like hallways) as public, while a greater de- gree of privacy is accorded within the common areas of hospitals and prisons which are regarded as “semi-pri- vate.” 10 The expectation of privacy is also informed to a great extent by the person(s) affected. The Omnibus Budget Reconciliation Act (OBRA) lists amongst its nursing home resident rights, “the right to privacy with regard to accommodations, medical treatment, written and telephonic communications, visits and meetings of family and of resident groups.”^11 The procedures of the CareMedia project would clearly impinge upon these rights if the recordings took place in the absence of informed consent from the residents or their legally authorized representatives. Nursing home employees, in contrast, appear to be entitled to limited expectation of privacy. The Omni- bus Crime Control and Safe Streets Act of 1968, also known as the Federal Wiretap Act, permits silent video surveillance without the need for consent, and audio recording when at least one party to the conversa- tion has given consent, though certain states require consent of all parties involved. 12 Legal scholars have argued that this Act protects employers from prosecu-
Stakeholder
Residents
Professional Staff (e.g., physicians, nurses)
Non-professional Staff (e.g., housekeeping)
Nursing Home Administration
Families and Visitors
Oversight Agencies (e.g., state surveyors)
Payors (e.g., Medicare, Medicaid)
Staff Unions
Clinical Researchers
Table 1
Ethics of Video Surveillance in Nursing Homes: Stakeholders and their Interests
Interests
- Quality-of-Life (QoL)
- Quality-of-Care (QoC)
- Privacy
- Confidentiality
- Freedom from abuse and neglect
- Reliable, valid data to guide assess- ment and treatment
- Confidentiality of medical records
- Privacy with residents
- Protection from malpractice or wrongful allegations
- Privacy
- Protection from wrongful allegations
- Maximize QoL and QoC for residents
- Minimize legal risks to facility
- Optimize fiscal health of facility
- Privacy
- Optimize QoL and QoC for loved one
- Reliable, valid QoL and QoC data upon which to base (re-)certification
- Investigate complaints promptly, aided by surveillance data
- Documentation of quality measures as is done by the Nursing Home Compare website
- Estimation of effort of caring for resi- dents, and thus appropriate payments to nursing facilities
- QoL and QoC consistent with state and federal guidelines
- Optimize work conditions
- Minimize risk of malpractice or wrongful accusations
- Reliable, valid data collection and analysis
- Appreciation of care processes
- Appreciation of ethical challenges
- Appreciation of advantages and limi- tations of technology in real-world settings
- Ability to conduct research with ad- equate safeguards while accommodat- ing minority dissent
race & ethnicity • fall 2006 615
Bharucha, London, Barnard, Wactlar, Dew, and Reynolds
tion when the expressed purpose of the surveillance is supervision or evaluation of the employees. 13 Moreover, continued employment in the context of prior notifica- tion of electronic surveillance technologies within the facility implies informed consent, and thus would fur- ther diminish the staff ’s reasonable expectation of pri- vacy. Still others have asserted the distinction between privacy rights for professional versus nonprofessional staff, lobbying for greater physical and informational privacy for professional staff. None of the above argu- ments, however, would likely extend to spaces such as staff locker rooms which would be considered private.
A more problematic issue is the extent to which pri- vacy rights apply to visitors. The courts have been di- vided in their opinions regarding the privacy rights of “casual, transient visitors,” basing their determinations primarily on the nature of the relationship between the nursing home host and the visitor, and the area of the facility within which the privacy right is asserted.^14
Privacy: The Moral Perspective The very organization of this discussion into a legal and a moral perspective highlights the fundamental conceptual and semantic complexities of capturing all that is subsumed under the rubric of privacy rights. Prominent scholars have dealt with this issue by defin- ing it variously as a value or a moral claim, or as a legal right.^15 From the moral perspective, privacy is viewed as a prerequisite for the pursuit of personhood, free- dom and responsibility. This deontological perspec- tive dates back to antiquity when both the Greeks and Romans recognized and acknowledged the public ( res publicae ) and private ( res privatae ) spheres of life. As Allen argues, the emphasis in these ancient cultures was to idealize and celebrate the domain of public, community engagement, while dismissing the private world as mundane; in contrast, modernity places the highest moral value on the private sphere while accom- modating public impingements for the sake of collec- tive welfare.^16 Seminal to the ethical deliberations about the Care- Media project are the premises of the recently articu- lated interest theory of rights.^17 According to this per- spective, the normative force for, and justification of a
right emanates from the interests of those that the right is intended to protect or facilitate. The right to privacy, from this vantage point, is viewed as protecting and advancing the ability of affected parties to retain effec- tive control over a zone of intimacy that encompasses their person and aspects of their private lives and con- duct. Having effective control over such a zone of inti- macy is essential for individuals’ ability to preserve the personally meaningful ways in which they signify and differentiate various grades of personal relationships from different kinds of more formal relationships, and public roles and persona from their private lives. One advantage of the interest the- ory of rights is that it recognizes that the rights of individuals can come into conflict. LTC facilities represent a con- text in which such conflicts frequently exist. For example, simply because residents are demented or debilitated does not mean that they have lost all interest in retaining a sphere of inti- macy surrounding their personal and private life. In fact, because the ability of such persons to retain effective control over this zone of privacy is often compromised by their dementia or other medi- cal conditions, they frequently require the assistance of caregivers in retaining the integrity of this sphere of privacy. At the same time, however, LTC facility staff have obligations to look after the welfare interests of residents, and these obligations frequently require in- trusions into their intimate spaces. As a result, LTC facilities already have to balance the provision of a safe and commodious environment, in which the health and welfare interests are supported and advanced, with the need to support residents in retaining as much con- trol as possible over a legitimate zone of intimacy. The interest theory of rights seeks to resolve such conflicts by attending to the underlying interests of in- dividuals that such rights serve to secure, and searching for resolutions that are maximally responsive to each of these underlying interests. It is our contention that the CareMedia project should be understood within the larger context of conflicting rights that already exists in many LTC environments. Although the CareMedia project creates additional intrusions into the privacy of residents, it is our contention that this intrusion, and the specific ways in which it was both targeted and limited, can be justified by the important ways in which this technology is responsive to the underlying interests that generate the rights in question.
Privacy Protections Although the technology employed within CareMedia may be viewed as in tension with, or intruding upon the
The interest theory of rights seeks to resolve such
conflicts by attending to the underlying interests
of individuals that such rights serve to secure,
and searching for resolutions that are maximally
responsive to each of these underlying interests.
race & ethnicity • fall 2006 617
Bharucha, London, Barnard, Wactlar, Dew, and Reynolds
Since staff and visitors are also being recorded, HIPAA would appear to have limited relevance in the context of our research. Moreover, our interest lies as much in the capture of “patient” information as it does in “health” information, which is the substance of HIPAA. Neither the legal mandate nor the spirit of HIPAA appears to be well-positioned to encompass affect, behaviors, activity patterns and social interac- tions which are the gist of CareMedia. If the purpose of confidentiality is to de-identify and protect “sensitive” information, who determines the sensitivity of and ac- cess to the information? Under what conditions can the data be compelled to be released? Historically, fed- eral certificates of confidentiality have been available to protect human-subjects research involving potentially stigmatizing or sensitive issues; however, these also apply only to the subjects themselves and do not nec- essarily protect the research facility, or other partici- pants who are not research subjects per se. Without the protection offered by a mechanism such as the federal certificates of confidentiality, it is understandable that research facilities would fear legal consequences, and hence abstain from research activities without which there is little hope of addressing chronic plaguing clini- cal problems.
Confidentiality Protections Protection of confidentiality is a formidable challenge since we collect personally identifiable information in the form of facial and body images and voices. We typi- cally offer the option of obscuring faces and voices to non-participants, but in extreme situations, we have the capability to delete individuals from scenes en- tirely. The latter process is, however, considerably more labor-intensive, costly, and counterproductive to our goal of examining social interactions. Faces and voices of all study participants are de-identified in any presen- tation of data outside the immediate research team. Some potential study participants have raised the objection that the de-identification process takes place only after the filming is completed, and fear that third parties may gain access to the recordings before non- participants are de-identified. The longer-term goal is to develop technologies that will automatically re- cord (in real-time) only those who are identified by the computers as study participants. In the interim, data are protected by a firewall and the added security of a Virtual Private Network (VPN). Essentially, only those investigators, who have two separate passwords, and a key to the locked data processing lab, would be able to access the information, and the data are not subject to exploitation by internet hackers. Ultimately, we envision contemporaneous data analysis so as to obviate the need to store raw video and audio data;
all that will remain will be the derived metadata. Such an advance will require that a data safety monitoring board is actively involved in verifying that the results of our investigations are not based on fraudulent raw data that are immediately discarded and therefore not available for further scientific inspection.
Informed Consent
Educational Component Since the video and audio recordings are not limited specifically to study subjects, we were ethically obli- gated to educate all individuals who may have access to the dementia unit regarding the objectives and pro- cedures of the study. In order to honor “cold calling” policies for research, the nursing home administration initially mailed an introductory letter on their letter- head with a simply worded two-page CareMedia study announcement to the legally authorized representa- tives of the dementia unit residents for their review. The legal proxies were encouraged to contact either the nursing home administration and/or the research team with any questions or concerns. Within two weeks of this mailing, the families were telephoned by the de- mentia unit social worker to obtain their permission for the research team to establish contact with them directly. With one exception, all legally authorized representatives agreed to do so. Subsequently, several educational meetings were offered in the evenings to the families of the dementia unit residents. Typically, these meetings were poorly attended; however, fami- lies were very receptive to phone contact in compliance with “cold calling” policies. In addition to the general educational meetings for families, specific training ses- sions were scheduled with all staff members who would have access to the dementia unit during the study pe- riod. Upon completion of this educational component, families were notified via letter of the date of study ini- tiation four weeks ahead of time. Moreover, a sign was placed on the entrance of the dementia unit two weeks prior to and throughout the study period informing all who entered of the title, objectives, and procedures of the study as well as the name and contact information for the principal investigator and the clinical research associate. They were specifically informed of the film- ing in the shared spaces of the dementia unit, and the location of space(s) that were not being filmed in order to provide them more privacy.
Subject Assent An attempt was made to obtain affirmative agreement for study participation from all residents, noting any verbal or nonverbal communication of dissent as dis- approval for study participation. Due to the advanced stages of their neurodegenerative illness, no resident
618 journal of law, medicine & ethics
INDEPENDENT
was able to demonstrate even the most basic under- standing or appreciation of the fact that we would be filming on the dementia unit for research purposes, the voluntary nature of their participation, or its implica- tions for their well-being. We took the added precau- tion of ascertaining from the nursing home caregivers whether the timing of our visit was “just a bad day” for the resident, or whether they truly were very cogni- tively limited, precluding meaningful assent. Support- ing our judgment that these residents lacked capacity for assent or dissent was the fact that a durable power of attorney for health care made treatment decisions in every case. The components of truly valid assent/ dissent, and the procedures implemented to obtain it from severely cognitively impaired potential research subjects are important areas for future research.
Informed Consent Process with Legally Authorized Representatives All legally authorized representatives, with one excep- tion, provided proxy informed consent for their loved one to participate in our research, with the clear under- standing that there would be no direct benefits from this research to their loved one (“non-therapeutic re- search”). It is important to clarify, however, that these individuals were legally authorized to make clinical decisions on behalf of their loved ones, not necessarily research enrollment decisions. None of our study sub- jects had specifically-stated, written documentation of advance research directives. It is therefore unclear to what extent the proxy informed consent represented a substituted judgment on behalf of the study subject. As with the need to determine the components of valid as- sent/dissent, there is considerable national debate per- taining to the need to examine how legally authorized representatives for research initiatives are identified and appointed, and the nature of their deliberations about enrolling their loved one in a research project.^18 A major ethical obstacle that has previously arisen during the implementation of our research at another site merits deeper reflection: a resident, his/her legal proxy or staff member not only declines study par- ticipation but objects to any recordings taking place on the unit whatsoever. In contrast to traditional clinical research which typically concerns the rights of study participants only, our study captures many non-par- ticipants whose interests must be proactively protected as well. We attempted to do so by posting the afore- mentioned study notice on the dementia unit entrance two weeks prior to and throughout the study period. Legally authorized representatives of our study sub- jects were encouraged to inform regular visitors to the dementia unit of the nature and inclusive dates of our study. Additionally, the Commonwealth of Pennsyl-
vania’s Institutional Review Board advised us to keep a large, common activity space un-instrumented as a compromise in order to partially accommodate those who may be opposed to recordings taking place on the unit. The broader ethical tension here is balancing and respecting the refusal of a minority of potential subjects and/or their legally authorized representatives with the overwhelming majority of individuals who choose to participate in the study, and the benefits that may result from our findings for future LTC residents with dementia.
Informed Consent with Staff Members No staff member objected to the recordings on the unit as long as they had assurance that their faces and voices would be de-identified. The nursing home adminis- tration had agreed to accommodate non-consenting staff member(s) on another unit in the facility without compromising care on the study unit. Staff members were also informed that physical or sexual abuse of a resident was the only condition that we were required to report to the nursing home administration. More- over, the nursing home would not have access to any of the filmings, and there would be no punitive ac- tions against staff members for study participation or refusal. Local agencies which supply nursing staff on an as-needed basis were informed ahead of time by the nursing home administration of our study so as to provide only those individuals with no serious objec- tions to the study procedures.
Future Directions
CareMedia challenges and potentially expands the dia- logue pertaining not only to the ethics of surveillance technology research, but also to the widening gulf be- tween individual-centered and public health ethics. We have demonstrated the application of the principles of the interest theory of rights to address the critical issues of privacy and confidentiality protection in an innovative project that films the lives of cognitively impaired nursing home residents and their caregivers continuously in real-time. In addition to the strategies we had to develop to address the principal ethical is- sues, an immediate lesson for us has been the need for LTC facility administrations to carefully consider and articulate a research mission statement that, like the clinical mission statement, would be reviewed with the resident and his/her legally authorized representative prior to admission to the facility. Ideally, the agreement to abide by this statement should be obtained in writ- ing. While no generic mission statement could possibly hope to cover specific details of potential projects, the residents and/or their legally authorized representa- tives should have a clear understanding that this facil-
