Partner notification in the context of HIV: an interest-analysis

One of the things that learned professions such as law, the clergy, medicine, and
even public health have in common is codes of confidentiality. These codes ensure
the sanctity of the intimate conversations that characterize them. In public health
practice for example, client-provider confidentiality, has been treasured for ages
1]. The earliest explicit codification of confidentiality in the context of medicine
is attributed to Hippocrates of Cos (460 BC – 370 BC). Titled “Oath of Hippocrates”
the code prescribes rewards to the secret keeper, for living up to his obligations
and punishments for those who do not (see extract below):

“Whatsoever in the course of practice I see or hear (or even outside my practice in
social intercourse) that ought never to be published abroad, I will not divulge, but
consider such things to be holy secrets. “Now if I keep this oath and break it not,
may I enjoy honor, in my life and art, among all men for all time; but if I transgress
and forswear myself, may the opposite befall me…”2].

Millennia upon millennia, medical oaths continue to agree fundamentally with the tenets
of the Hippocratic ethics. This is not to say that professionals’ obligation to keep
the secrets of their clients has been polemics-free. Alan Brandt in his “No Magic Bullet, the social history of venereal diseases in the United States …” animates the vehemence with which physicians within the United States (US), and those
from across the Atlantic debated the issue of partner notification 3]. Brandt recounts the various medical, military, and public health responses that
have arisen over the years – including the incarceration of prostitutes during World
War I to the establishment of the extensively debated premarital blood tests. Of note,
Brandt calls for the destabilization of the naturalness with which efforts that combat
sexually transmissible infections have centered on punishment for sexual misconduct.

Introduced for syphilis and then extended to include gonorrhea in the 1930s and 1940s
3],4], partner notification is now considered useful for a wide range of sexually transmissible
infections (STIs) 5]. Partner notification is currently defined as the process whereby the sexual partner(s)
of a case (an index patient) is/(are) identified and informed of their exposure, then
invited to testing, counseling and, where necessary, treatment 6],7]. As a term, “partner notification” has undergone some metamorphoses in the past decade
and-a-half. Prior to 1998, guidelines of the US Centers for Disease Control and Prevention
(CDC) contained “contact tracing” or “partner notification”. These terms were dropped
in 1998 for “partner counseling and referral services” 8]. The current CDC guidelines use “partner services” 6]. Related terminologies in current use include “expedited partner therapy”, “privilege
to warn” “duty to warn” and “duty to disclose”. While many academics consider these
as equivalent activities, public health practitioners consider them to be vastly different
in most respects except for the principal goal 9]. Each has unique logistic and substantive issues. For instance, partner notification
as is currently defined, requires that only sexual partners be notified that they
have been in contact with a sexually transmissible infection. It is a confidential
process: details of the index cases are known only to the health professionals treating
them and are not divulged either to sexual partners or to disease notification systems.
Contact tracing entails all the health care’s actions taken to trace and actively
and systematically contact all the partners/contacts indicated by the index person
as having had relationships with him/her at risk. Bayer and Toomey discriminate between
contact tracing and duty to warn 10]. They note that the moral “duty to warn”, arose out of the clinical setting in which
the physician knew the identity of the person deemed to be at risk. This approach
provided a warrant for disclosure to endangered persons without the consent of the
patient and could involve the revelation of the identity of the “threatening” party
(the index patient). Bayer and Toomey lament that confusion regarding the approaches
has led many to mischaracterize processes that are fundamentally voluntary as mandatory,
and processes that respect confidentiality as invasive of privacy.

We note, however, that the terminology of partner notification differs between countries;
it is used interchangeably with contract tracing in such countries as the United Kingdom
and Australia. The global guidelines from the World Health Organization (WHO) 11] use partner notification. The WHO and the Joint United Nations Programme on HIV/AIDS
(UNAIDS) recommend that partner notification be done on a voluntary basis 12]. Where partner notification has to be done without consent, the UNAIDS provides clear
guidelines 13] such as: (i) The HIV-positive person in question has been thoroughly counselled;
(ii) Counseling of the HIV-positive person has failed to achieve appropriate behavioral
changes; (iii) The HIV-positive person has refused to notify, or consent to the notification
of his/her partner(s); (iv) A real risk of HIV transmission to the partner(s) exists;
(v) The HIV-positive person is given reasonable advance notice; (vi) The identity
of the HIV-positive person is concealed from the partner(s), if this is possible in
practice; (vii) Follow-up is provided to ensure support to those involved, as necessary.

As illustrative examples, we summarize below the global outlook of partner notification
using relevant resources from the US, Europe, as well as the WHO global guidelines.
These guidelines provide background and basis for the roll out of partner notification
services globally. First, a technical report commissioned by the European Centre for
Disease Prevention and Control (ECDC) details the public health benefits and context
for partner notification within Europe 5]. The report notes that, some countries have wide-ranging legal obligations to enforce
partner notification, others have laws that are not enforced; and there are countries
in the region with no such laws at all. There are laws or regulations in eleven European
countries that make partner notification compulsory for the healthcare provider, the
patient or both 5].

Second, the CDC recommendations for partner services for Human Immunodeficiency Virus
(HIV) infection and three other STIs (syphilis, gonorrhea, and chlamydial infection)
provide implementation guidance in the United States 6]. As used in the recommendations, “partner services” is broad, with partner notification
as a critical component. Functions of partner services include prevention counseling,
HIV and STD testing, treatment or linkage to medical care, linkage or referral to
other services.

Third, the WHO in April 2012 issued guidance on couples HIV testing and counseling
including antiretroviral therapy (ART) for treatment and prevention in sero-discordant
couples 11]. Partner notification is a prominent feature of the guidelines.

Beyond this background, we present a review of the principles and goals of partner
notification, partner notification methods and approaches, and the data behind partner
notification.

Review

We constructed searches with relevant key words in OvidMedline, and Embase. These were supplemented with ‘unrestricted searches” in Google Scholar. Citations
of retrieved resources were also perused for additional resources. A careful study
of these and other resources provided the background data for our analysis. The first
search in OvidMedline with key words [HIV infections, partner notification or contact
tracing, privacy] spanned the period 1946 to week 4 of September 2013 and yield 83
records. These were manually reviewed and reduced to 27. A related search in the same
database replaced partner notification or contact tracing, with partner service or
expedited partner therapy and yielded 27 records that were manually reviewed and reduced
to six. A third search in Embase using same key words yield two resources. Both were
relevant.

Principles and goals of partner notification

Cognizant of the hypothesis that secrecy nurtures disease by providing an environment
conducive to the spread of infection, Gostin and Hodge 14] note that one of the earliest recorded public health strategies for STI prevention
was to pierce the veil of secrecy surrounding these hidden diseases by notifying sexual
partners of infected patients. This was supported by the moral theory that sexual
partners could take precautions and seek medical treatment if the risk of infection
was disclosed. Once the risks of infection were identified, the incidence of STD infection
would decline suggestively as infected persons reduced behaviors that placed them
at risk for disease 12]. The following principles as set out in all partner notification guidelines, serve
as the foundation for providing partner notification services: Such services ought
to be client-centered, confidential, voluntary, free, evidence-based, comprehensive
and integrative, culturally, and developmentally appropriate, accessible and available
to all 5],6].

Every partner notification program seeks to maximize access to HIV prevention services
by providing all infected persons (index patients) with support to ensure that the
partners are confidentially informed of their exposure, tested, and linked to medical
care. At the community level, partner notification seeks to reduce future rates of
transmission in line with the current treatment as prevention arguments 15]-17]. By linking partners to care and reducing their viral load, the community viral load/level
of infectiousness is reduced 18]-22]. Also, by identifying persons with previously undiagnosed HIV infection through partner
notification and linking them to medical care services, and possibly to ART, partner
notification services help reduce transmission within the community.

Partner notification methods and approaches

Conventionally, four strategies have been used in implementing partner notification:
provider referral, self-referral, contract referral, and dual referral.

Provider referral notification involves a partner being notified of their possible
exposure by a health professional who has been specifically trained to locate and
notify partners. The healthcare worker then links the partners to medical, prevention,
and support services while protecting the confidentiality of the index patient.

In a patient referral, the index patient takes responsibility for informing their
sexual partner(s) of their possible exposure to an STI and for referring them to services.
Introduced in the 1970s in response to high levels of gonococcal infection 23], patient referral has since been widely used for a wide range of STIs 5],6],24].

Contract referral notification involves index patients selecting specific partners
they prefer to notify and agreeing to a specific time frame in which they will do
so. Patients agree that if they do not notify the selected partners within the established
time frame, the provider will notify the partners. Dual referral notification involves
an index patient and the provider (or third party) jointly notifying a partner of
exposure 5]. The data in support of partner notification are summarized below.

The data behind partner notification’ in the context of HIV

Credible data exist in support of partner notification’s role in HIV prevention-care
cascade. We present some of these data. In the US for example, when syphilis prevalence
peaked in 1990, partner notification among other national public health interventions
contributed to its reduction and, later, elimination 25]. The efforts were adjudged to be cost efficient 25],26]. Also, evaluation of 10 years of data from the New York program for notification
and referral services for gonorrhea, as well as of other program data showed a reduction
in gonorrhea prevalence 27],28].

Other calls for the implementation of partner notification exist. For instance, it
is argued that, of the approximately 1–1.2 million persons living with HIV infection
in the United States, 25% are not aware of their infection; transmission from persons
not aware of their infection accounts for 54%-70% of new infections 29],30]. A case is therefore made for partner notification given that it effectively identifies
persons with previously undiagnosed HIV infection 31].

We present now the interests and counter interests in partner notification.

The interests in partner notification

Partner notification as a public health intervention, is implemented by governments
to, among other reasons, reduce the incidence of new cases of STIs in individuals
and in the community 5]. However, such interests of the government sometimes conflict with those of the individual.
Two cases in the United States, the Jew Ho v. Williamson (1900) and Jacobson v. Massachusetts
(1905) speak to such conflicting interests in public health practice. In his distillation
of Jacobson v. Massachusetts, Gostin notes that governmental interests, though grounded
firmly in statutory directives such as the police power^a, are not absolute 32]. Gostin clarifies that police powers cannot be exercised in a manner that violates
the constitutional rights of individuals.

Resolution of these conflicted interests may call for legal jurisprudence, intervention
by ethicists/bioethicists, or both. “Interest Analysis” has been applied in the past
to resolve related conflicting interests 33],34]. Coined by Brainerd Currie in 1963, and popularized by Gostin and Hodge 14], interest analysis, as we will apply in this paper will entail a disinterested analysis
of the various interests in partner notification, and not in legal jurisprudence.
The interests at play here are the interests of the index patient, the interests of
partner, and the interest of the public health profession/al. These interests are
discussed.

Public health interests in partner notification

The public health interests of partner notification are justified in a number of ways.
First, such interests are premised on the theory that partners are entitled to knowledge.
This theory posits that “knowledge empowers individuals to avoid continuing risks;
knowledge of infection allows for early treatment; and that knowledgeable partners
can adapt their behavior to prevent further transmission of infection to others” 14],35]. Available evidence shows that people are less likely to have unprotected sex once
they are aware of their HIV status 36],37]. Additionally, equipping public health officials with knowledge helps to prevent
infection of and/or to treat sexual partners – control the spread of STI and its attendant
consequences in the general public 35],36]. A review by Christie and Kendall show that prevalence of HIV among identified contacts
ranges from 15% to 30%, with the majority of contacts being unaware of their possible
exposure. Their conclusions concur with afore cited works 25]-28] that, that partner notification is an effective public health intervention 38]. Thus partner notification programs offer substantial benefits to three principal
groups: persons infected with HIV, their partners, and the community 6]. Presented this way, public health interest may be viewed as coinciding with those
of the index patient, the interests of his/her partner, and the interests of the community
as a whole.

Individual level interests

While governmental interests in partner notification may be equated to the interest
of all, persons infected with HIV have individualized or counter interests 14]. For instance, an infected person may desire to maintain the privacy of their health
status from unnecessary disclosure because of the negative impacts of disclosure (done
consensually or otherwise). Emotional or physical abuse by or against the index patient,
domestic violence of a physical or psychological nature as well as dissolution of
a long-standing relationship 39],40] have been associated with partner notification. Theresa Yuricic 41] writes about how stigma particularly within the African-American community potentially
decreases the efficacy of partner notification laws in this population. Disclosure
of HIV status she notes can result in social stigma among their family and friends.
Yuricic further discusses how disclosure makes them vulnerable to discrimination in
employment, housing, and insurance 41]. Angela Nicoletti 42] enumerates other interests including fear of losing family support. Negative impacts
of partner notification on infant feeding have been recording in Ghana 43]. It may thus seem reasonable, given these potential and real negative implications
on the index patient, to respect their interests when systems or health personnel
are unable to assure their security. In this regard, screening for likelihood of violence
and/or other serious negative consequences on the index patient ought to be an important
part of the partner notification process.

The bioethicist’s interests in partner notification

The interest of the bioethicist in this matter is to facilitate a moral resolution
of the conflicted interests. S/he is interested in providing nuanced answers to such
questions as whether or not index patients or healthcare providers have a moral duty
to inform partners of their exposure risk, whether or not the healthcare provider’s
relationship with the index patient takes priority over the obligation to protect
others from the patient’s infection, or just how important is the partner’s right
to know that he or she may be at risk? These are not ordinary questions. Gostin and
Hodge 14] note that while the law provides its own answers to these questions, ethical reasoning
or moral resolution is important. The fundamental conceptual problem is the difficulty
in telling which of the claims (the index patient’s or the partner’s) should take
precedence. Our analysis draws on Kass and Gielen’s 14],44] as well as Gostin and Hodge’s 14] earlier works. Like Kass and Gielen, we believe that principles-based analysis may
help in the bioethicist’s search for answers to the above questions.

Partner notification and the principle of beneficence

To Tom L. Beauchamp, and James Franklin Childress, beneficence means that persons
have the responsibility to do good for others, to prevent harm to others, or, at the
very least, to avoid directly harming others 45]. In the context of partner notification, promoting beneficence would mean carefully
balancing the harms and benefits of partner notification – balancing the various interests.
Without a doubt, any effort that seeks to prevent new HIV infections is pregnant with
beneficence. An important question that earlier commentators have asked is: “how ethically
effective partner notification programs are in achieving this goal” 44]. To provide a reasonable response as to the extent partner notification programs
satisfy the principle of beneficence, Kass and Gielen 44] suggest a set of five assumptions all of which need to be met. These are:

1. The index patients must know and be willing to disclose the names of their contacts,
and it must be possible to locate the contacts;

2. a significant proportion of the persons contacted must not have known already that
they had been exposed to HIV, and they must have been practicing unsafe practices
before and be willing to change to safer practices now;

3. A significant proportion of the persons contacted must be willing to be tested;

4. Some number of those found to be HIV-infected will not have known already that
they were infected;

5. Some number of persons found to be HIV-infected must have been practicing unsafe
practices before being contacted who will change to safer practices after being informed
of their potential exposure.

We suggest a sixth assumption – partners who willingly accept to be tested when found
to be HIV+ would have unfettered access to HIV treatment and prevention services for
self and other partners. We note however, that arguments in support of this newly
introduced assumption, needs to be nuanced by discussions on the ethical tensions
associated with HIV treatment commodity allocation dynamics. For instance in settings
where ARVs are very scarce, arguments concerning whether or not it is ethical to divert
scarce ARVs from PLHIV who meet both the implicit adherence and the explicit medical
eligibility criteria to others who do not. In support of the former, Elliot Marseille
and colleagues argue forcefully for the supremacy of prevention over treatment even
when it means denying treatment to medically eligible PLHIV 46]. Brock and Wikler 47] also argue that ‘the strongest moral imperative directs us to give priority to saving
the most lives even if this means lowering the priority given to the goal of universal
access to treatment, to provide maximum protection from HIV infection’. Macklin and
Cowan on the contrary argue that it will be unethical to ‘deliberately watch patients
with treatable AIDS worsen and die, if medications for treatment are diverted to for
this purpose 48].

The extent to which Kass and Gielen’s assumptions were met then or now is unknown.
There are no published studies evaluating partner notification programs on the terms
stated above. Studies whose findings could qualify as near approximations suggest
that partners who were notified were significantly more likely to modify their sexual
behavior and to avoid intercourse than those who had not been notified 49],50]. Other studies show that partners believe their chances for preserving their individual
health rely in substantial part on notification of risk 17],37]. It follows, they argue, that universal notification is a plus to preserving and
promoting public health. Yet, many infected persons, particularly women, counter these
observations by documenting the costs of partner notification in the form of domestic
violence and abuse 37],39],40].

Determining the balance between these competing interests is not easy. In its current
application, partner notification may not represent a proper balance – “beneficence-wise”.
This paper’s stance based on a conditional consequentialist reasoning supports the
roll out of partner notification. In line with current arguments for treatment as
prevention (TasP) and pre-exposure prophylaxis (PrEP), partner notification should
have an additional condition of ensuring that index patients and their notified partners
have access to HIV treatment and prevention commodities. The second condition (which
entails prior screening for possible partner abuse) is inked in the latest guidelines
by the CDC 6].

Partner notification and the principle of justice

Justice, according to Beauchamp and Childress, requires that people be treated fairly
45]. That is restrictions cannot be imposed on, or benefits provided to, one person or
one group of people when another similarly-situated person or group is treated differently
without adequate justification 45]. Any partner notification program that mirrors the above is morally indefensible.
For instance, programs that prioritize men who have sex with men (MSM), female sex
workers, or index patients suspected to have concurrent multiple sex partners or selectively
notifies their partners should have a sound scientific justification. Such sound justifications
would include for example, evidence showing that the group is significantly at greater
risk of transmitting or acquiring the infection – in the case of MSM and sex workers.
In that case disproportionate benefits/risks will not be unjust.

Per evidence 6],11], partner notification policies may be said to be just, but not necessarily practicable
or fair implementation-wise. Pottker-Fishel’s analysis of legal cases, state statutes
and other literature shows that many states’ statutes on partner notification are
ineffective and burdensome, placing too much responsibility on health care providers
51]. There are also other studies that suggest that physicians may execute their responsibilities
with regard to partner notification inconsistently depending on the demographics of
their HIV-infected patients. For instance, findings of a national survey of US physicians
that assessed among others partner management, and clinical practices for STIs including
HIV show that variables such as gender, race, and sexual orientation would influence
a decision to maintain or to violate patient confidentiality 52].

From the female index patient’s perspective, partner notification programs while beneficial
for women who are HIV-negative, may be harmful for HIV-positive women (costs such
as domestic violence and other abuses have been discussed in this paper). Some commentators
have argued that since uninfected women stand to benefit more from the implementation
of partner notification programs than infected women, partner notification may be
unjust. Second, given that women are at greater risk of heterosexual transmission
than are men, it may be argued that heterosexual women are more likely than heterosexual
men to benefit from partner notification. These are relevant justice questions. We
nevertheless, believe that the disproportionate benefits in these instances are not
unjust, as they are also disproportionate risks.

Partner notification and the principle of respect for autonomy

The third bioethical principle of respect for autonomy, means that people must be
respected as autonomous agents who have the right to make decisions for themselves
without interference from others 45]. Some commentators have argued that, this principle creates a moral right to decide
whether, when, and to whom one should release personal information. On the one hand,
if this principle is followed to the letter, and in the context of partner notification,
an index patient in his/her quest to preserve his/her privacy can simply choose not
to disclose his/her partners’ names. On the other hand, a partner’s normative claim
about invasion of autonomy is also reasonable. Partners, according to Gostin and Hodge
14] can appeal to autonomy in claiming a right to know. Partners could argue that they
cannot make rational, or autonomous choices in the absence of relevant information.
Indeed, Sarah Conly’s recent work titled “Against Autonomy: Justifying Coercive Paternalism”53] would seem to favor the partner’s claim. The autonomous index patient’s right to
engage in behaviors of his/her choosing does not extend to behaviors that can result
in serious harm to partner(s). On the basis of these, an index patient may have no
legitimate moral claim to maintain his/her confidentiality when doing so has a real
potential to harm (HIV infection and its associated consequences in this case).