… another research notebook entry.
Some of the psychiatric readmissions literature considers “readmissions” after very long intervals, like a year or two … or ten (!!!). I wonder if these are properly called “readmissions.” Suppose I have bones more prone to fracture than is typical. Suppose I am treated for a fracture this year and for another one a year from now. Is the second a readmission? I don’t think anybody would consider it so. Other than the common factor of my condition, the second break has nothing to do with the first. These are just two admissions.
(Yes, one could cook up an example where the second and first are related by other modifiable factors, but I think you get my point. Sometimes similar injuries befall someone for reasons that cannot be helped. That’s more likely the larger time window examined. Maybe I see my doctor every year or two for respiratory illness. Are those re-visits or just separate visits?)
Similarly, suppose I have a mental health condition that flares up and requires hospitalization in 2012 and again in a year. Is the second a readmission? Or is it just a new admission? What if it is two, three, or ten years later?
Readmissions are normally examined within a much tighter interval, like 30, 60, or 90 days. They might be indicative of suboptimal care during or after the index admission. An assertion that they are is less credible the longer the interval.
I say all this as a warning that some of what is termed “readmissions” in the psych literature perhaps should not be so named.
A second warning: I’m noticing some relatively recent publications on psychiatric readmissions using some very old data. Lastly, some of the work is non-U.S. focused. There’s nothing wrong with that, but it may not be generalizable to U.S. settings.
From “Factors associated with multiple re-admission to a psychiatric hospital,” by Webb, Yaguez, and Langdon:
Leaving these methodological concerns aside, there is however, a degree of concordance between some studies in terms of the factors associated with re-admission. These factors include: (a) diagnosis of a psychotic illness (Bernardo & Forchuck, 2001; Daniels, Kirkby, Hay, Mowry, & Jones, 1998; Hodgson, Lewis, & Boardman, 2001; Kastrup, 1987; Lewis, & Joyce, 1990; Korkelia, Lehtinen, Tuori, & Helenius, 1998; Rabinovitz et al., 1995), (b) being young (Kastrup, 1987; Langdon et al., 2001; Lewis & Joyce, 1990; Vogel & Huguelet, 1997; Woogh, 1986), (c) being male (Appleby et al., 1996; Haywood, Kravitz, Grossman, & Cavanaugh, 1995; Kastrup, 1987; Korkelia et al., 1998; Lewis & Joyce, 1990), (d) being divorced or unmarried (Bernardo & Forchuck, 2001; Hodgson et al., 2001; Rabinovitz et al., 1995), (e) substance misuse (Haywood et al., 1995; Langdon et al., 2001; Woogh, 1986), (f) greater symptom severity regardless of diagnosis (Swett, 1995; Postrado & Lehman, 1995; Lyons et al., 1997), (g) unemployment (Bernardo & Forchuck, 2001; Haywood et al., 1995; Rabinovitz et al., 1995), (h) mode of admission (Hodgson et al., 2001; Korkelia et al., 1998; Vogel & Huguelet, 1997), (i) higher level of education (Bernardo & Forchuck, 2001; Rabinovitz et al., 1995), (j) non-compliance with medication (Haywood et al., 1995; Weiden & Glazer, 1997), (k) quality of life (Postrado & Lehman, 1995), and (l) disruptive behaviours (Sullivan, Young, & Morgenstern, 1997). However, there is also a degree of inconsistency between studies with respect to some factors being associated with re-admission. For example, some studies have reported that ‘‘revolvingdoor’’ patients tend to be younger (Kastrup, 1987; Korkelia et al., 1998; Langdon et al., 2001; Lewis & Joyce, 1990; Vogel & Huguelet, 1997; Woogh, 1986) while others have reported that ‘‘revolving-door’’ patients tend to be older (Haywood et al., 1995; Rabinovitz et al., 1995). Others still have suggested that there may be a relationship between sex and diagnosis which may affect re-admission rates (Daniels et al., 1998; Lewis & Joyce, 1990; Vogel & Huguelet, 1997).
- The study is of patients in the U.K., using data from 1999.
- “Those patients who were found to have had three or more admissions within their lifetime were defined as the RD [revolving door] group.”
Several variables were shown to predict membership in the ‘‘revolving-door’’ group and findings replicate Langdon et al. (2001), although there were differences. ‘‘Revolving-door’’ patients may have more enduring and chronic mental illnesses, but were similar to their ‘‘non-revolving door’’ counterparts on some variables. Research of this nature is difficult given the cross-sectional nature of studies, and a lack of a clear consensus within the literature as to which factors are associated with ‘‘revolving-door’’ service users remains.
From “A Case-Control Study of Factors Associated With Multiple Psychiatric Readmissions,” by Silva, Bassani, and Palazzo:
Several studies have identified strong predictors of readmission, such as poor treatment adherence (9–13), low level of education (11,14), deficient follow- up after hospital discharge (14), involuntary admission (3,15), lack of social or family support (11,16), and diagnoses of schizophrenia (17) and substance use disorders (10,18). [^^^]*
Individuals [in Brazil] who had been referred to community psychosocial support units after their most recent discharge had about 20% lower odds of multiple readmissions than those referred to usual outpatient care. Those who lived closer to the hospital (residents of the same city) were more likely to have multiple readmissions. The adjusted multivariate hierarchical analysis revealed that a diagnosis of schizophrenia or psychotic symptoms was associated with multiple readmissions, as were younger age at first admission and a greater number of previous admissions.
From “Rehospitalizations Among Psychiatric Patients Whose First Admission was Involuntary: A 10-Year Follow-Up,” (PDF) by Paula Rosca:
Objectives: To examine the characteristics of patients whose first admission to Israel’s psychiatric units was involuntary, and to identify a specific profile of the patients at increased risk for future readmissions. Our hypothesis was that when the first admission of a patient was involuntary, the number and duration of future hospitalizations would be greater. Method:We used information extracted from the National Psychiatric Case Registry on all patients admitted for the first time during 1991 (N=2,150) and on their follow-up over the next ten years. Chi-square statistics were used to test for significance differences in demographic and clinic variables between patients hospitalized voluntarily and those hospitalized involuntarily. Multiple regression analysis was performed to identify a specific profile of risk of recidivism during a ten-year period (1991-2000). Results: Compared with patients admitted voluntarily, those who were admitted involuntarily had a significantly greater number and duration of rehospitalizations. They were more likely be diagnosed as suffering from schizophrenia while the voluntarily admitted patients were more likely be diagnosed as having an affective disorder. Risk factors for the number of readmissions included: young age, legal status (involuntary) of the first admission, as well as period of residence in the country. Risk factors affecting the duration of readmissions were single/widowed status, native born and a suicide attempt in the twomonths prior to the first admission. Conclusions: Two distinct profiles for the number of readmissions and inpatient days were identified. A diagnosis of schizophrenia and selected demographic variables were better at predicting risk of recidivism than the involuntary legal status of the first admission. [Emphasis added.]
From “One year outcome in first episode schizophrenia,” by Üçok, Polat, Çakır, and Genç:
- “2002). Recent studies reported that duration of untreated psychosis (DUP) has an independent contribution to outcome throughout first year of the illness (Larsen et al.2000;Malla et al.2002;Harrigan et al.2003). Longer DUP was found to be associated with more severe negative and/or positive symptoms at 12 months (Harrigan et al. 2000).”
- “Noncompliance to the medication is one of the most common causes of psychotic relapse and rehospitalization in patients with schizophrenia.”
- “Male gender appears to be another possible predictor of poor outcome.”
- “The presence of subjective depressive feelings during the first admission was associated with earlier relapse while the presence of depressive delusions and higher educational attainment protected against early relapse (Geddes et al. 1994).”
- “In this study we investigated the predictors of relapse and one year outcome in a homogenous group of patients with first episode schizophrenia. Considering the multi-dimensionality of the outcome concept presence of relapse and rehospitalization, level of global functioning, employment status and severity of symptoms at one year were analyzed.Premorbid social functioning in early childhood and compliance to the treatment were independently contributed to occurrence of relapse as hypothesized. Contrary to our expectations, the results of this study did not support the importance of DUP and severity of symptoms as predictors of one year outcome in first episode schizophrenia.”
* I just made up the “[^^^]” notation. It’s just like “[...]” but it means that the following passage comes earlier, not later, in the document from which I quote.