Tuesday, November 17, 2015

Beliefs That Dictate Evidence: Open Visitation in the ICU (Again)

The cart belongs behind the horse.
I recently blogged on idealogues who haven't any interest in the truth, rather their interest is in defending their beliefs.  For these true believers, evidence is sought selectively and strength of belief is not apportioned to strength of evidence.  Beliefs reign supreme, and evidence serves the beliefs.  The cart leads the horse.

And so let it be with open visitation in the ICU.  I'm interested in this because it is an issue of practical concern for me, and my interest was recently piqued because in nursing school, my wife was taught that open visitation is better for everyone and that ample evidence supported this contention.  Today, I came across a tweet about ICU visitation policies, a statement from the American Association of Critical Care Nurses.  So I decided to investigate a bit further the evidence upon which their policy proposals are predicated.

The very first statement in the "Supporting Evidence" section of this document is "In practice, 78% of ICU nurses in adult critical care units prefer unrestricted policies."  This statement is at odds with my personal experience working with ICU nurses for the better part of the past 20 years.  While they are patient and family advocates generally, they also recognize that the exigencies of the ICU environment require some limitation of visitation, and so does their own psychological well-being.  So I began by investigating references 7-13 which are proffered in support of this statement which for some (many?) lacks face validity.  Here are those seven references, a description, and a synopsis taken from the abstract of each:

Reference 7:  Am J Crit Care. 1997;6(3):210-217.  Current Practices Regarding Visitation Policies in Critical Care Units.  This is a descriptive study, a survey of 201 nurses done in 1997.  It says: 
  • "Seventy percent of official policies for visitation were restrictive. In practice, 78% of nurses were nonrestrictive in their visitation practices. Variables that affected practices regarding visiting hours were the patient's need for rest, the nurse's workload, and the beneficial effects of visitation on patients. Requests from patients and their families were ranked least important.  Restricted hours were perceived to decrease noise (83%) and promote patients' rest (85%). Open visitation practices were perceived to beneficially affect the patient (67%) and the patient's family (88%) and to decrease anxiety (64%). Perceptions of ideal visiting hours included restrictions on the number of visitors (75%), hours (57%), visits by children (55%), and duration of visits (54%), but no restriction on visitation by immediate family members (60%). "  (Emphases mine.)
The 201 nurses surveyed in this study perceived a need for balance between the needs of patients, the nurses, and patients' families.  That 78% of nurses were unrestrictive in their policies, at their discretion, while working in a restrictive setting does not mean that 78% of them "prefer unrestricted policies."  That 99% of the time I wear my helmet when I ride a motorcycle does not mean that I favor laws that require me to do so.  This interpretation of this study is flatly disingenuous.
Reference 8:  J Adv Nurs 2009;65(11):2293-2298 .  The impact of hospitals visiting hours policies on pediatric and adult patients and their visitors.  This is a [systematic] review article.  There is no new evidence here.
Reference 9:  Crit Care Med. 2007 Feb;35(2):605-22.  Another review.  However, this one states in its conclusion that they're making recommendations in spite of the lack of credible, high quality evidence:
  • "More than 300 related studies were reviewed. However, the level of evidence in most cases is at Cochrane level 4 or 5, indicating the need for further research."
Reference 10: Dimens Crit Care Nurse 1998; 17(1) 40-47.  Visitation:  Policy Versus Practice.  I can't even find this reference, except in the other references.
Reference 11:  J Nurs Care Qual. 2001 Jan;15(2):18-26.   Visitation in critical care: processes and outcomes of a performance improvement initiative.  This is a review of selected literature followed by a change in visitation at one hospital and a subsequent survey.
 Reference 12:   2005 Jan-Mar;40(1):18-28.  Visiting hours in the ICU: finding the balance among patient, visitor and staff needs.  This is a qualitative study consisting of interviews with eight nurses who "were considered experts".  Perhaps the most honest of all the references, its conclusion is summarized in the title.
Reference 13:   2006 Jan;15(1):13-27.  American Association of Critical-Care Nurses' national survey of facilities and units providing critical care.  This is a descriptive survey study with a poor response rate conducted by the same organization that promulgates open visitation as a divine edict.  It is not even a study of visitation - that was just one of many broad ranging questions the entire survey addressed.  The data on visitation is presented as a descriptive table in Figure 8 which adds little to the debate.  This reference does not support the statement to which it is originally attached.

Now, this task of policing this pseudoscience has already taken me two hours of my afternoon and I don't have the time or the patience to adjudicate the relevance and integrity of the rest of the 51 references, but I suspect that this is about as good as they can do to support with evidence their belief in how visitation ought to be.

So, as is all too common, we have a situation where the belief has primacy and the evidence is subordinate.  The evidence is seen as a tool to defend the belief which is a foregone conclusion.  And this is maddening, absolutely maddening, to have the cart before the horse like this.

Rational people develop beliefs based upon evidence, and the strength of their beliefs is proportional to the evidence.  That is not at all what is happening here.  This document is a piece of propaganda to convince the unsuspecting of the validity of the agenda they have laid out.

This does not mean, however, that the agenda is wrong.  It only means that robust evidence does not support the agenda.  And maybe it need not.  This is after all a social policy engineering enterprise, and we often don't need, or can't get credible evidence to inform social policies.

Instead what we must do is determine all the "stakeholders" (I hate buzzwords, but it seems an apt one), ask them what they value, and make the inevitable compromises to balance the benefits to all.  My concern is that this agenda, perhaps driven by "patient satisfaction scores" and associated incentives, is not trying to balance the values of all stakeholders (nurses being one key stakeholder group), and that it is using smoke and mirrors claims of "evidence" to obfuscate the agenda.

ETA:  More thoughts today pondering  this with myself and considering Facebook comments (why does everybody comment there but none here, he asked with forlorn rhetoric?)

  1. Why the absolutism?  Nobody is arguing against visitation, they are arguing for limits or compromise.  When did compromise fall out of favor, especially among progressives?
  2. "Evidence", especially weak or dubious evidence, or claims of evidence that does not exist should NOT be used to stifle debate, or shield against it, or to silence the voices of dissension.
  3. How ironic that groups that are so beholden to the strictest standards of evidence are willing to loosen their standards when it suits their means and their ends.
  4. Responses to surveys are highly dependent upon how you ask the questions.
  5. If families don't understand the reasons why limitations are being imposed, they are likely to advocate for unrestricted visitation, thinking it costs nobody anything.  If the questions were posed to them acknowledging the costs and trade-offs for others, they, being good people, would probably we willing to accept restrictions, and accommodate the competing needs of others.  Refusal to consider this unnecessarily paints patients' families as inflexible and uncompromising.


  1. I completely support a cluster randomized trial. But given that today, ICUs must make some sort of decision regarding visitation, what does the preponderance of evidence (while limited) support in your examination of the literature?

  2. I've been thinking about this for several days. I think there are some problems that cannot be solved using "evidence based methodology" - there's an article in the NEJM today about "evidence based end of life care" - is that really a topic on which we can made decisions with evidence? Some evidence is too weak to base decision upon, so we use common sense. I am most strongly against absolute capitulation to the desires of families in this and other regards. I disagree that serving them is the end-all be-all. So I think we should recognize that they want liberal visitation, but that RNs and MDs, for various reasons, need some limits on it. Or, I need a closed office in the ICU so I cannot be hovered over by the son of bed XYZ while I dictate another patient's chart in the "Doc Box." Or, maybe the RNs need two 15-minute coffee breaks that happen like clockwork and a half hour lunch that is scheduled and for which they have floating relief. We should take a holistic look at all of the problems and make logical common sense compromises to achieve a balance in all of our goals.

  3. I have it on confidence that at least some of the nurses in the "studies" used to support open visitation do not feel as though they are at liberty to disagree with the avant-garde which is pushing this "agenda". If that is so, then the entire research enterprise may represent one giant Hawthorne Effect demonstration. Agenda yields policy yields "research" to support the former. Junk science.


Note: Only a member of this blog may post a comment.