From the Editor
Some patients are so ill that we take away their basic rights and freedoms, admitting them involuntarily to hospital. But how common is the practice?
In the first selection, we consider a new paper by Michael Lebenbaum et al. that looks at involuntary admissions from 2009 to 2013. They find the percentage is not only high (by international standards) but that it has soared in recent years – from 70.7% in 2009 to 77.1% in 2013.
We consider a recent essay on AI in the second selection. Google has made international headlines with its program, Duplex, that can call and book appointments. In this piece, the authors note that AI has failed to live up to its potential. “Schedule hair salon appointments? The dream of artificial intelligence was supposed to be grander than this…”
DG
Hospitalizations and Involuntary Admissions
“Prevalence and predictors of involuntary psychiatric hospital admissions in Ontario, Canada: a population-based linked administrative database study”
Michael Lebenbaum, Maria Chiu, Simone Vigod, Paul Kurdyak
British Journal of Psychiatry Open, March 2018 (Open Access)
Involuntary psychiatric hospital admissions generally occur when an individual with mental illness is admitted to hospital against their will owing to a perceived imminent danger to the individual or others and unwillingness to remain in hospital voluntarily. These admissions are not desirable, because they can be disruptive to the patient–provider relationship and have a negative impact on the patient’s perception of their care at the time of the admission and afterwards. Although there is variability across jurisdictions, involuntary admissions are common and increasing in high-income countries, including several European countries such as the UK, The Netherlands and Germany. There are few studies investigating trends in involuntary admissions over time in other jurisdictions.
Involuntary admission to hospital may, to some extent, be an avoidable event if care provided in community settings mitigates psychiatric crises that precipitate involuntary admissions. Studies mostly from Europe have shown that, in addition to these service-related factors, risk factors for involuntary admissions include patient sociodemographic characteristics such as being an immigrant or ethnic minority, and clinical severity variables such as a diagnosis of a psychotic disorder. Studies from other jurisdictions are largely from the USA and emphasise the importance of available mental health resources. Canada, the most recent country-wide prevalence estimate of involuntary admissions (25%) was in the late 1970s. One recent study among 200 patients with first-episode psychosis at four sites in Ontario had a prevalence of involuntary admissions of 68.6%, suggesting a large increase in prevalence. However, no recent population-based Canadian study has examined prevalence of involuntary admissions among all patients. Furthermore, existing Canadian studies have examined only a very limited set of characteristics of involuntarily admitted patients. This is consistent with many international studies, which are often small in size, based on one or a few sites, and examine only a limited number of risk factors. The aims of this study were to determine the trends in prevalence of involuntary admissions in a large population-based North American sample covering the whole Province of Ontario, Canada (population ~14 million), and to examine the independent risk factors that predict involuntary admission status in this jurisdiction.”
Michael Lebenbaum
Here’s what they did:
- “We examined the prevalence and risk factors of involuntary admissions among all patients admitted to mental health and addictions (MHA) beds in Ontario, Canada. Annual prevalence was examined from fiscal year (April 1 to March 31) 2009 to 2013, and risk factors for involuntary admission were examined for all years pooled.”
- They drew from Ontario data (the Ontario Mental Health Reporting System, as well as several other databases for demographic information).
- An involuntary admission was defined as either a Form 1 or Form 3 present at admission.
- They looked at demographic characteristics, including age and gender, as well as prior mental health use, and clinical variables, like diagnoses and depressive symptoms.
- Statistical analyses were done, including a trend over time for the prevalence of involuntary admissions, and modified Poisson regression to determine the unadjusted and multivariable adjusted associations between risk factors and involuntary admissions
Here’s what they found:
- “A total of 250 773 admissions were identified between fiscal years 2009–2013 among adults living in Ontario with valid health card numbers.” Those with inadequate data and forensic cases were excluded, leaving a final sample of 115 515 individuals.
- “Among the 115 515 individuals who met inclusion criteria, 85 607 (74.1%) were involuntarily admitted. The prevalence of involuntary admissions significantly increased from 70.7% in 2009 to 77.1% in 2013 (RR = 1.021 per year)… 28 726 (33.6%) of individuals who were involuntarily admitted were released within 72 h of admission.” See figure below.
- Factors linked to the likelihood of involuntary admission: past week police contact (RR = 1.36) and severity variables including high-acuity triage status (RR = 1.55) and medium triage status (RR = 1.30). Factors less linked: self-harm identified in the emergency department contact preceding the admission (RR = 1.17), not having a psychotic disorder (RR = 0.83–0.93, depending on the diagnosis), and having a mental health visit in the week preceding admission with either a family physician (RR = 0.89) or a psychiatrist (RR = 0.94).
- Also (using the adjusted model): younger age (RR = 1.10), immigrant status (RR = 1.07) and previous involuntary assessments/admissions (RR = 1.09) all increased the likelihood of involuntary admission.
- See the figure below for a Forest plot of unadjusted and fully adjusted associations between clinical scales and involuntary admissions.
In this large population-based study, we found a very high prevalence of involuntary psychiatric admissions – almost three-quarters of all hospital admissions. The prevalence has increased steadily over the years between 2009 and 2013, from 70.7 to 77.1%. There were also significant risk factors for involuntary admission to hospital among sociodemographic, past service utilisation, pathway to care, and clinical severity characteristics. The sociodemographic factors included immigration status and young age, both of which increased involuntary admissions. Mental health service utilisation in the prior year was also a strong predictor, with physician contact reducing the likelihood of involuntary admission. Police contacts in the week prior to admission, a pathway to care variable, and a number of severity variables including triage status, self-harm and psychosis were also risk factors increasing the likelihood of involuntary admissions.
A few thoughts:
- This is a good study.
- The data is impressive – drawing on Ontario hospitalizations, and including more than 115,000 people.
- The trend is clear. Wow. The percentage of involuntary admissions is way up.
- How does this compare to other countries? “Reviews of mostly European countries, covering the period 1971–2000, found that the prevalence ranged from a low of 1.0% in Spain to a high of 93% in some hospitals in Switzerland, with all other estimates less than 50%. Therefore, the prevalence found in our study is among the highest reported prevalences for a Western country.”
- Why are there more involuntary admissions? The authors offer a few explanations, including the lingering effects of deinstitutionalization. “Psychiatric beds in Ontario dropped from 219 to 81 per 100 000 population from 1965 to 1980, and further dropped to 34.2 per 100 000 population in 2015, a decline of 84.4% over the 50-year period. The reduction in the number of hospital beds may have resulted in the remaining beds being preferentially deployed for the most severely ill presentations, which would be highly correlated with the need for involuntary admission.”
- Is physician compensation a factor? Forms pay well. Too well?
AI and the Future
“A.I. Is Harder Than You Think”
Gary Marcus and Ernest Davis
The New York Times, 18 May 2018
https://www.nytimes.com/2018/05/18/opinion/artificial-intelligence-challenges.html
The field of artificial intelligence doesn’t lack for ambition. In January, Google’s chief executive, Sundar Pichai, claimed in an interview that A.I. ‘is more profound than, I dunno, electricity or fire.’
Day-to-day developments, though, are more mundane. Last week, Mr. Pichai stood onstage in front of a cheering audience and proudly showed a video in which a new Google program, Google Duplex, made a phone call and scheduled a hair salon appointment. The program performed those tasks well enough that a human at the other end of the call didn’t suspect she was talking to a computer.
Assuming the demonstration is legitimate, that’s an impressive (if somewhat creepy) accomplishment. But Google Duplex is not the advance toward meaningful A.I. that many people seem to think.
If you read Google’s public statement about Google Duplex, you’ll discover that the initial scope of the project is surprisingly limited. It encompasses just three tasks: helping users ‘make restaurant reservations, schedule hair salon appointments, and get holiday hours.’
Schedule hair salon appointments? The dream of artificial intelligence was supposed to be grander than this — to help revolutionize medicine, say, or to produce trustworthy robot helpers for the home.
Gary Marcus
So begins a short essay by New York University Professors Gary Marcus and Ernest Davis. The essay is short, highly readable, and doesn’t require much of a summary.
They note with pessimism the lack of advancement in the field. “The reason is that the field of A.I. doesn’t yet have a clue how to do any better.”
Marcus and Davis focus on Google Duplex. They make a few points:
- “As Google concedes, the trick to making Google Duplex work was to limit it to ‘closed domains,” or highly constrained types of data (like conversations about making hair salon appointments), ‘which are narrow enough to explore extensively.’”
- “The crux of the problem is that the field of artificial intelligence has not come to grips with the infinite complexity of language.”
- “Open-ended conversation on a wide range of topics is nowhere in sight.”
Though there is potential in machine learning, the authors argue that the field will always be limited. “No matter how much data you have and how many patterns you discern, your data will never match the creativity of human beings or the fluidity of the real world. The universe of possible sentences is too complex. There is no end to the variety of life — or to the ways in which we can talk about that variety.”
A few thoughts:
- This is a good essay.
- You can listen to Google Duplex in action and draw your own conclusions; see: https://www.youtube.com/watch?v=bd1mEm2Fy08. (On a tangent, it’s clearly Laurel not Yanny.)
- The second selection in the Reading of the Week is meant to be lighter. Is this piece lightweight? Maybe not. While it is clever, it also touches on a larger problem: the challenges of computers trying to interact with us – and the fact that language use is so varied and diverse. (Think of trying to define and describe sarcasm.)
- Is our experiment with AI a bust? Marcus and Davis certainly feel it is, closing:
Today’s dominant approach to A.I. has not worked out. Yes, some remarkable applications have been built from it, including Google Translate and Google Duplex. But the limitations of these applications as a form of intelligence should be a wake-up call. If machine learning and big data can’t get us any further than a restaurant reservation, even in the hands of the world’s most capable A.I. company, it is time to reconsider that strategy.
In an era of self-driving cars, are they overstating their point? AI has many branches, and the complexity of language makes the task they describe especially challenging (and so, no need for those of us in mental health to re-train just yet).
Reading of the Week. Every week I pick articles and papers from the world of Psychiatry.
Recent Comments