- Email: [email protected]
Predictors of institutionalization for people with multiple sclerosis
ARTICLE IN PRESS
Disability and Health Journal
-
(2015)
-
www.disabilityandhealthjnl.com
Research Paper
Predictors of institutionalization for people with multiple sclerosis Lilian U. Thorpe, M.D., Ph.D.a,b,*, Katherine Knox, M.D.c, Rochelle Jalbert, B.Sc.d, June Hyun-Ja Lim, Ph.D.a, Darren Nickel, Ph.D.c, and Walter J. Hader, M.D., F.R.C.P.c a
Department of Community Health and Epidemiology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada b Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan, Canada c Department of Physical Medicine and Rehabilitation, University of Saskatchewan, Saskatoon, Saskatchewan, Canada d College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
Abstract Background: Multiple Sclerosis (MS) is a chronic, progressive disease of the central nervous system with a high prevalence in Canada. While the disease course is highly variable, a significant portion of people with MS may spend more than 10 years living with severe disability, and many of those will eventually require full time institutional care. Despite the high personal and economic cost of this care, little is known about predictors of institutionalization. Objective: The objective of this study was to identify predictors of institutionalization. Methods: Longitudinal data from a university MS clinic database were extracted to explore nursing home placement over time of an urban subgroup. Cox regression analysis was performed with age of MS onset and sex, as well as baseline information obtained at the first MS clinic assessment: MS course, Kurtzke Expanded Disability Status Scale score, and functional system scores. Results: Older age of onset ( p 5 .019) and higher baseline scores in specific functional systems (cerebellar, bowel/bladder, brainstem, and cerebral/mental) were significant ( p 5 .000, p 5 .000, p 5 .001, p 5 .000 respectively) predictors of nursing home placement. Conclusions: Patients with older age of MS onset and those with baseline impairment in specific functional systems (cerebellar, bowel/ bladder, brainstem, and cerebral/mental) may be at higher risk for future institutionalization and should be assessed with particular care to determine potential avenues of support to minimize this. Ó 2015 Elsevier Inc. All rights reserved. Keywords: Multiple sclerosis; Institutionalization; Nursing facilities; Prognosis
Multiple sclerosis (MS) is a chronic degenerative, inflammatory disease of the central nervous system with onset of first symptoms typically in early adulthood. MS survival rates may be increasing, leaving persons with MS living well into older age,1e3 often with significant disability. There is considerable variability in the literature on disability outcomes, with median time to needing a cane between 15 and 32 years after disease onset.4 A relapsingremitting MS (RRMS) course at onset is most common,
Financial disclosure: The authors have no conflicts of interest to declare. This research was supported by a University of Saskatchewan Deans summer student award (project number: Med-12-25). Presentation of abstract at meeting: A poster ‘‘Predictors of Institutionalization for People with Multiple Sclerosis’’ has been accepted for presentation by one of the authors (RJ) at the Fifth Cooperative Meeting of the Consortium of Multiple Sclerosis Centers (CMSC) and Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) in Dallas, Texas from May 28e31, 2014. * Corresponding author. Department of Community Health and Epidemiology, University of Saskatchewan, 107 Wiggins Road, Saskatoon, Saskatchewan, Canada S7N 5E5. Tel.: þ1 3069667977. E-mail address: [email protected] (L.U. Thorpe). 1936-6574/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2014.10.002
offering a more favorable initial prognosis compared to the 10% presenting with a primary progressive MS (PPMS) course at onset. However, 50% of those with RRMS convert to a secondary progressive MS (SPMS) course an average of 15 years after first MS symptoms.5 Progress has been made in the treatment of RRMS, although the impact on long term disability outcomes remains a subject of controversy and no effective treatments are yet available for progressive forms of the disease.6 More information is needed on predicting and planning the care needs of the aging MS population. It has been estimated that over 93,500 people in Canada are living with MS.7 There is considerable geographic variation in prevalence rates across the country.8 Saskatoon, within the province of Saskatchewan, has been reported by Hader & Yee to have a particularly high prevalence rate of 293.8 per 100,000 with a yearly incidence of about 9.5/ 100,000 per year.9 This high prevalence, coupled with the natural features of the disease predict the need for high levels of personal and nursing support related to MS care. Unfortunately, despite the progressive and variable nature of the disease course, there is little literature exploring risk
ARTICLE IN PRESS 2
L.U. Thorpe et al. / Disability and Health Journal
factors for institutionalization specific to MS. Data from the US National Multiple Sclerosis Society (cited by Buchanan, Wang & Ju, 2002) suggested that 20e25% of people with MS eventually end up in long term care, and that the risk might be particularly high in those with bowel dysfunction, poorer health and greater functional decline, as well as in those with greater caregiver burden.10 Studies based on other populations have found that bowel system dysfunction11,12 and cerebellar and cortical system atrophy13 may be increased in institutionalized individuals. A review of general community-based studies14 suggests that age, low self-rated health, functional and cognitive impairment, dementia, prior nursing home placement, and high number of prescriptions are risk factors for nursing home placement. Institutionalization is likely to depend not only on individual needs, but also on the availability and proximity of resources. Saskatchewan is a province large in geography but small in population, which results in people being dispersed over large distances, often removed from the major sources of formal support. This might be expected to result in higher institutionalization rates in general (and for MS patients in particular), and indeed, Saskatchewan was reported to have the second highest ratio of longterm care beds per population in Canada in 2008 at 112.8 beds per 1000 population aged 75þ.15 Another factor potentially contributing to institutionalization is insufficient planning for future care needs. Putnam & Tang suggested that most people with MS are not adequately informed about the availability of services for possible future needs, and frequently do not plan adequately for long term care.16 Even if sufficient alternate community-based supports are available, a lack of readiness to discuss services or knowledge about services could lead to early caregiver burnout and precipitous institutionalization. The Multiple Sclerosis Society of Canada states that individuals themselves generally ‘‘prefer to live in the comfort of their own homes.’’17 In spite of that, little information is available concerning the prevalence of institutionalization or predictors for institutionalization in general or specific to the Canadian MS population. Institutionalization removes people from their familiar environment and family supports, and may decrease their overall autonomy and quality of life. Social isolation within the nursing home is a particular issue for those with MS as they are usually younger and more mentally alert than the general, usually elderly population.18 Health may also be adversely impacted by institutionalization, as home-based care (compared to institutional care) has been associated with improvements in various health dimensions.19 Lastly, institutionalization is very costly to the health care system, with hospitals and other institutions taking up the largest portion of the nation’s expenditures.20 For all of these reasons, it is essential to improve the understanding of the prevalence of and contributors to
-
(2015)
-
institutionalization. While the natural history of MS and risk factors for disease progression in treated and untreated populations have been evaluated,6 these studies generally focus on mobility (i.e., needing a cane to walk), relapse rates, imaging or more recently, quality of life outcomes. Predictors related to institutionalization outcomes identified at clinical assessments might help focus resources optimally to minimize or delay institutionalization. The objective of this study was to identify predictors of institutionalization, using the longitudinal data from a well-established longitudinal MS clinical database maintained at the University of Saskatchewan. We chose to focus on the functional system sub-scores of the EDSS as these sub-scores provide a graded rating of impairment and related function.
Methods The clinical database was initially established as a registry by one of the authors (WH) in 1969 at which time records of patients with MS were extensively searched. Sources included the Saskatoon MS clinic (located in one of the acute care hospitals in Saskatoon), other physicians, nursing homes or home care, and the three city hospitals. Patients were also identified through contact with the MS society and through other provincial records.9 This database was updated every five years under the guidance of WH until 1990 as well as in 1986, 1996 and 2003. Clinical cases coming through the MS clinic at any time or through any of the three Saskatoon hospitals between 2001 and 2005 were added or had their information updated. Yearly phone calls ascertained residence type and living status. The entire group was most recently updated in December 2011 and the records of patients living in Saskatoon at the last contact were additionally updated in summer 2012. Publicly published obituaries were regularly searched by the clinic administrator to update the registry when directly obtained data were missing. The Saskatoon MS Clinic registry as of August 30, 2012 included 1217 patients, of whom 150 had been followed since 1977.21 Standard diagnostic information in the MS registry included clinically-definite, probable, and suspected MS; sex; place of birth; date of birth; place of onset of MS; age at diagnosis of MS; ethnic origin; family history; and date of death. Most records also included disability levels according to Expanded Disability Status Scale (EDSS) scores, functional system subscales,22 and course of MS (relapsing-remitting, secondary progressive, and primary progressive) at assessment times. The MS Clinic registry had additional information available including marital status, education, religion, physical comorbidities, measures of mental health, specific symptoms such as bladder problems, seizure history, events preceding diagnosis, smoking history and income status. Data chosen for our analysis were those with good validity (for example, standardized
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
measurements), consistency and completeness (information gathered consistently over the course of patient follow up period). The protocol for the study was developed collaboratively with the research and clinical team, and ethics approval was obtained from the University of Saskatchewan Biomedical Research Ethics Board. Analyses were conducted on deidentified data. All persons listed in the Saskatoon MS Clinic registry who had an individual, initial assessment at the Saskatoon MS Clinic for the first time prior to 2010, confirming clinically-definite MS by the diagnostic criteria applicable at the time,23e26 and who resided in the Saskatoon Health Region at last contact were included in the initial data selection. The 2010 McDonald diagnostic criteria were not applicable as they were published in 2011, after the cutoff date for study inclusion. When information was incomplete, this was augmented from the clinical record under the supervision of clinicians involved with the project. Records were left in the analysis if there were baseline EDSS and functional systems data and if patients were not already in a nursing home when they were first assessed at the MS clinic. Information collected for this analysis included date of birth, sex, date of onset of first MS symptoms (MS onset), date of initial clinic assessment, course of MS, most recent living situation (own home, own apartment or condominium, assisted living, personal care home, nursing home, other institution or other), first date of permanent admission to a nursing home, date of death and disability level. We use the term nursing home in our document to refer to residential facilities providing 24-h, supervised institutional long-term care services to meet the needs of individuals, usually having heavy needs that cannot appropriately be met in the community. This type of care is formally termed a special-care home by our Ministry of Health. Course of MS included relapsing-remitting, primary progressive, secondary progressive and benign MS. This latter course is most frequently described as minimal disability ten years after onset. After prolonged follow up however, the majority of those with relapsing-remitting course convert to a secondary progressive MS course. MS course at the first clinic assessment was utilized for the analysis as this was determined to be the earliest and most reliable course classification available from the registry. Disability was measured with the Expanded Disability Status Scale (EDSS), the most commonly utilized scale rating neurological impairment in MS. The EDSS is an ordinal scale ranging from a score of zero (no disability) to ten (death due to MS). The EDSS is especially influenced by mobility restrictions and is also calculated according to the subscores from the functional systems: pyramidal, cerebellar, brainstem, sensory, bowel/bladder, visual and cerebral/ mental. Scores for the named functional systems range from zero (normal) to maximum scores of five (cerebellar, brain stem, cerebral/mental) or six (pyramidal, sensory,
-
(2015)
-
3
bowel/bladder, visual), with maximum scores indicating severe impairments in each. Cerebral impairment includes mood alteration only, memory loss, or dementia. Brainstem impairments may include nystagmus, dysarthria, and dysphagia. Cerebellar impairments include coordination deficits. SPSS version 21 was used to perform the data analysis. Data were explored, charted and tabularized in a descriptive format to assess suitability for further statistical analysis. Based on extensive exploration of data distribution and association with outcome variables, cerebral functional system scores (range 0e3) were recoded into a new categorical variable (<2 5 low and score 3 5 high) and brainstem functional system scores (range 0e4) were recoded into a new categorical variable (<2 5 low and scores 3e4 5 high). Available and reliable clinical and demographic variables were entered into a univariate Cox regression model to assess their effects and magnitude of impact on the outcome. The start date for the Cox regression was taken to be the date of the first MS clinic assessment, as baseline assessments with potential predictive value were reliably available from this date. The date of onset of first MS symptoms was generally only estimated by patients, which would result in recall bias. The end date was taken to be the date of first permanent (not respite) admission to a nursing home. Patients were censored if they were not in a nursing home at the last contact. Variables were then added into a multivariate Cox regression model and removed sequentially if their impact was not at a significance of p ! .05. Sex and age of MS onset were left in the model regardless of significance in accordance with established procedures.
Results Of the 1217 patients listed in the Saskatoon MS Clinic registry who had at least one individual assessment at the Saskatoon MS Clinic, 817 had been assessed in the clinic for the first time before January 1, 2010. Of these, 700 resided in the Saskatoon Health Region at the last contact, and 401 of those had both a baseline EDSS score and were not already residing in a nursing home at baseline. Demographics of the study sample are shown in Table 1. A total of 122 males and 279 females were included in this study. Consistent with selection criteria, the most recently enrolled participant was first assessed at the clinic in 2009. Mean follow-up time (years between the first clinic assessment and the last contact) was 13.2 (SD 7.9) years. The mean age at first assessment was 41.8 years (SD 11.6). Males seen in clinic were slightly older at the first clinic assessment than females [mean age 43.3 (SD 11.3) compared to 41.2 (SD 11.6)], but this difference was not statistically significant ( p 5 .85) using parametric statistics for the comparison of means (both male and female distributions were normally distributed in terms of age). The
ARTICLE IN PRESS 4
L.U. Thorpe et al. / Disability and Health Journal
Table 1 Demographics of the study population Male (%) Female (%) Mean age at MS onset (SD) Mean age at first visit (SD) Mean follow up in years (SD) Course of MS at first clinic visit Benign Primary progressive Progressive relapsing Relapsing remitting Secondary progressive Unknown or missing
279 122 31.9 41.8 13.2 N 12 53 7 207 120 2
(69.6) (30.4) (10.7) (11.6) (7.9) (%) (3.1) (13.2) (1.7) (51.6) (29.9) (0.5)
mean age of onset of MS symptoms was 31.9 (SD 10.7) years. Males also had a slightly older reported age of MS onset [mean age 33.3 (SD 10.6) compared to 31.2 (SD 10.7)], which was again not significant ( p 5 .79). At the first clinic visit, 12 patients were diagnosed with benign MS, 206 were diagnosed with relapsing-remitting MS, 120 were diagnosed with secondary progressive MS, 7 were diagnosed with progressive relapsing MS, 53 were diagnosed with primary progressive MS, and 3 had missing diagnoses. Disability scores (EDSS) at the first clinic visit were not significantly different between males and females using the ManneWhitney test for non-normally distributed variables ( p 5 .590). Median EDSS and functional system scores (with ranges in brackets) stratified by MS course are shown in Table 2. In general, patients with primary progressive and secondary progressive MS had the greatest deficits at the first clinic assessment. Fifty-six out a total of 401 (14.0%) were residing in a nursing home at the last contact. The greatest proportion (262/401, or 65.3%) still resided in their own home at the last contact. When stratified by MS type at first assessment we found that the greatest percentage living in a nursing home at the last assessment had been diagnosed with primary progressive MS at the first assessment. People with progressive relapsing MS at baseline had the youngest mean age at initial nursing home admission. However, the only significant predictor of younger age of nursing home admission was younger age of onset of MS. Table 3 shows the results of the multivariate Cox regression analysis for risk of nursing home admission. The final model included older age of MS onset and higher functional system scores at first clinic visit in cerebellar symptoms, bowel/bladder symptoms, brainstem symptoms and
-
(2015)
-
cerebral/mental symptoms as significant predictors of nursing home placement. The impact of gender and other variables initially in the model were not statistically significant. The impact of the significant predictors is illustrated graphically in Figs. 1, 2, and 3. In practical terms, every ten years older at initial assessment increased the hazard of nursing home placement by 1.4 times. Each increase of 1 in the score of baseline cerebellar symptoms increased the hazard of nursing home placement by 1.4 times. Scoring 3 compared to !3 in baseline cerebral-mental symptoms increased the hazard of nursing home placement by 8.7 times. Scoring 3e4 compared to !3 in baseline brainstem symptoms increased the hazard of nursing home placement by 4.7 times. An increase of 1 in the score of baseline bowel-bladder symptoms increased the hazard of nursing home placement by 1.4 times.
Discussion We found that 14% of patients seen in our clinic were admitted to a nursing home over a mean follow up period of just over 14 years (note that this rate is not the final institutionalization rate, for which we would have to follow each patient to death). Institutionalization rates for our participants may, however, actually be lower than rates found elsewhere for a variety of reasons including study participant selection factors, improved community or family supports, low availability of LTC beds in our health region, registry clinic selection bias, and possible non-random factors in data completeness of MS clinic patients. Our study also excluded persons with MS who were institutionalized at first assessment and therefore was likely to find a lower rate than that reported by cross-sectional study designs. Of future interest is a decrease in the LTC bed ratio over time (Peters, R, Saskatchewan Ministry of Health, personal communication, July 23, 2014) and an increasing acuity of patients in LTC over time (Ripley V, Saskatoon Health Region, personal communication, July 16, 2014), and it is unclear how this trend may affect the institutionalization of people with MS. Mental dysfunction was overwhelmingly the strongest predictor for institutionalization even though there were no persons with mental dysfunction in the severe range in our study. This finding is consistent with previous literature in suggesting that early cognitive impairment in MS is
Table 2 Median (range) EDSS and functional system scores by MS course at the first clinic visit MS course at first clinic visit EDSS Pyramidal Cerebellar Brainstem
Sensory
Bowel/bladder
Visual
Cerebral
Scale range Benign Primary progressive Progressive relapsing Relapsing remitting Secondary progressive
0e6 0.50 0.00 1.00 0.00 1.50
0e6 0.00 1.00 0.00 0.00 1.00
0e6 0.00 0.00 0.00 0.00 0.00
0e5 0.00 1.00 0.00 0.00 0.00
0e10 1.25 (0e4) 6.00 (0e9) 2.00 (0e4) 2.00 (0e9) 5.50 (0e9)
0e6 0.00 3.00 0.00 1.00 3.00
(0e1) (3e5) (0e3) (0e6) (0e6)
0e5 0.00 1.00 0.00 0.00 0.00
(0e1) (0e5) (0e2) (0e3) (0e5)
0e5 0.00 0.00 0.00 0.00 0.00
(0e2) (0e4) (0e1) (0e3) (0e4)
(0e3) (0e5) (0e4) (0e5) (0e5)
(0e1) (0e6) (0e6) (0e6) (0e6)
(0e1) (0e6) (0e5) (0e5) (0e6)
(0e1) (0e3) (0e2) (0e3) (0e3)
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
-
(2015)
-
5
Table 3 Cox regression analysis for nursing home placement by variables measured at first clinic visit Hazard ratio 95% CI p value Sex (reference: male) Age at onset of MS per decade Cerebellar system score at first clinic visit Cerebral-mental system score (3 vs. <2) at first clinic visit Brainstem system score (3e4 vs. <2) at first clinic visit Bowel-bladder system score at first clinic visit
1.078 1.379
0.591e1.966 1.055e1.80
0.806* 0.019
1.476
1.205e1.809
0.000
8.731
3.553e21.454
0.000
4.744
1.891e11.903
0.001
1.364
1.186e1.569
0.000
*Not significant but left in the model because of the potential significance of gender.
Fig. 2. Cox regression analysis for nursing home placement; impact of brainstem system score at first clinic visit.
predictive of later increased disability.27 While we do not know the degree of cognitive dysfunction among our participants at latest follow up, in the Alzheimer’s disease literature more severe cognitive impairment has been found to predict the loss of independent functioning. In our study, overall disability as measured by the EDSS and the functional system scores related to vision, other sensory and pyramidal involvement was not predictive of institutionalization. This differential impact of impairments related to specific systems may be explained by their differing impact on independent functioning. Some impairments may be mediated to a greater extent by the use of adaptive technologies. For example, a person with markedly restricted strength in their lower limbs (severe pyramidal system involvement) may manage their activities of daily living independently at a wheelchair level in their own accessible home. In comparison, a severely ataxic individual (severe cerebellar system involvement) who is unable to transfer without assistance, an individual who
has lost bladder or bowel function, or a person frequently aspirating (severe brainstem system involvement) may each have greater challenges in maintaining a safe, independent living environment. Future work should consider the functional impact of the system impairments in MS in predicting longer term independence. Enhanced care, services and technology may reduce the functional impact of these impairments in order to allow more people to stay in their own home. Older age of onset was also a significant predictor of future institutionalization; consistent with the literature that older age is associated with a shorter time interval to disability milestones.28 We did not find that the specific course of MS at first clinic assessment was predictive of institutionalization. This might be explained by the fact that well over half our sample included progressive forms of MS. At our baseline assessment time point many patients had already experienced substantial disability progression.
Fig. 1. Cox regression analysis for nursing home placement; impact of cerebral-mental system score at first clinic visit.
Fig. 3. Cox regression analysis for nursing home placement; impact of age at onset of MS symptoms.
ARTICLE IN PRESS 6
L.U. Thorpe et al. / Disability and Health Journal
Natural history studies have shown that disability progression rates are similar between relapsing and progressive MS after walking is restricted (EDSS 4).29 More recent work also demonstrates that disability rates are similar irrespective of initial MS course once moderate disability in even one functional system has developed (ESDD 3).30 Given our sample demographics and the natural history of later disability progression in MS, it is not surprising that we did not find a statistically significant differential impact of MS course on institutionalization in our analysis. Further research including patients assessed at earlier time points in their disease course might better clarify the differential impacts of various factors, including disease course on institutionalization. This information could provide practical benefits in terms of future planning for patients and their families.
Limitations The most significant limitation of our study involved incomplete information concerning various factors known to be relevant to institutionalization. For example, social support, socioeconomic status, and caregiver burden are important variables which were not consistently collected and therefore not included in our analysis. We were also not able to include individual changes over time in type of living situation (other than confirmation of nursing home placement) because of data over-writing and inconsistencies in the definitions of living situations. Since community residential supports as well as previous institutionalization and admissions to hospital may also be predictive of future long term nursing home placement,31,32 this would be important to capture in the future. We were also unable to include medical and other comorbidities in our analysis as they were not reliably collected in standardized format at the first clinic assessment. This was certainly a limitation for our analysis as one would assume (based on community analyses cited earlier14) that people who moved to nursing homes during our study period were more likely than those who remained in the community to have had comorbid medical and other problems. Another limitation in our study was change in diagnostic patterns over time, related to progress in imaging technology and changes in standard diagnostic criteria. For example, recent revisions to the McDonald criteria33 are thought to enable earlier diagnosis of MS, and changes like this would likely have had an impact on times to eventual institutionalization. We also did not have accurate information on the disease course at MS onset, and had to rely on information obtained at the first clinic assessment. The earlier diagnosis of MS over time could have led to potential cohort changes over time in this longitudinal study. There are also other factors which could have led to cohort changes over time including changes in the referral pattern to the clinic, and the impact of disease-modifying therapies
-
(2015)
-
which were not publically funded in the province of Saskatchewan until 1997. By the first clinic assessment, a significant proportion was already following a secondary progressive MS course, which decreased the utility of survival analysis. The later first presentation to clinic may be in part explained by a referral bias since the Saskatoon MS clinic is based within the Department of Physical Medicine and Rehabilitation. A referral bias to the rehabilitation-based MS clinic may also have impacted institutionalization outcomes. While persons seeking consultation at a rehabilitation-based MS clinic may have more advanced disease, they may also be more actively seeking resources to plan for the future effectively in order to delay or avoid institutionalization. Finally, our results are limited to those residing in the Saskatoon Health Region and those leaving the Saskatoon Health Region over the course of the follow up were not included. This is an important limitation especially since access to community home care supports and long term care access are variable across the province. Future work should include people living initially in rural settings where supports and services for independent living may be different than in urban centers.
Conclusion The final Cox regression analysis found that older age of MS onset and deficits in specific functional systems (cerebellar, bowel/bladder, brainstem, and cerebral/mental) at first clinic assessment were significant predictors of nursing home placement. Particularly important was the contribution of the cerebral/mental functional system score to institutionalization. This suggests that MS patients with these features should be assessed with particular care to determine potential avenues of support to maintain their ability to live independently. In addition to considering institutionalization as an important outcome, there is a need to prospectively collect standardized information about potentially predictive factors related to institutionalization in MS, such as medical comorbidities, self-rated health, prescriptions and psychosocial determinants. On a national level, this may be more feasible in the future through the establishment of a Canadian MS Monitoring System by the Canadian Institute for Health information.34 A more complete understanding of the relative contributions of a broader spectrum of risk factors related to institutionalization should help in developing strategies to keep people with MS in their own homes for longer.
Acknowledgments This study was supported by the College of Medicine, 2012 Dean’s Summer Student Research Projects, University of Saskatchewan. Research resources and space were
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
provided by the Department of Physical Medicine and Rehabilitation, Saskatoon Health Region.
References 1. Bronnum-Hansen H, Koch-Henriksen N, Stenager E. Trends in survival and cause of death in Danish patients with multiple sclerosis. Brain. 2004 Apr;127(Pt 4):844e850. 2. Grytten Torkildsen N, Lie SA, Aarseth JH, et al. Survival and cause of death in multiple sclerosis: results from a 50-year follow-up in Western Norway. Mult Scler. 2008 Nov;14(9):1191e1198. 3. Ragonese P, Aridon P, Salemi G, et al. Mortality in multiple sclerosis: a review. Eur J Neurol. 2008 Feb;15(2):123e127. 4. Tremlett H, Zhao Y, Rieckmann P, Hutchinson M. New perspectives in the natural history of multiple sclerosis. Neurology. 2010 Jun;74(24): 2004e2015. 5. Pittock SJ, Mayr WT, McClelland RL, et al. Disability profile of MS did not change over 10 years in a population-based prevalence cohort. Neurology. 2004 Feb;62(4):601e606. 6. Freedman MS, Selchen D, Arnold DL, et al. Treatment optimization in MS: Canadian MS Working Group updated recommendations. Can J Neurol Sci. 2013 May;40(3):307e323. 7. Statistics Canada. Table 105-1300-Neurological conditions, by age group and sex, household population aged 0 and over, 2010/2011, occasional (number unless otherwise noted), CANSIM (database). http:// www5.statcan.gc.ca/cansim/a26?lang5eng&retrLang5eng&id51051 300&paSer5&pattern5&stByVal51&p151&p2531&tabMode5 dataTable&csid5; Updated September 19, 2013. Accessed 19.12.13. 8. Poppe AY, Wolfson C, Zhu B. Prevalence of multiple sclerosis in Canada: a systematic review. Can J Neurol Sci. 2008 Nov;35(5):593e601. 9. Hader WJ, Yee IM. Incidence and prevalence of multiple sclerosis in Saskatoon, Saskatchewan. Neurology. 2007 Sep 18;69(12): 1224e1229. 10. Buchanan RJ, Wang S, Ju H. Analyses of the minimum data set: comparisons of nursing home residents with multiple sclerosis to other nursing home residents. Mult Scler. 2002 Dec;8(6):512e522. 11. AlAmeel T, Andrew MK, MacKnight C. The association of fecal incontinence with institutionalization and mortality in older adults. Am J Gastroenerol. 2010 Aug;105(8):1830e1834. 12. Carton H, Loos R, Pacolet J, et al. A quantitative study of unpaid caregiving in multiple sclerosis. Mult Scler. 2000 Aug;6(4):274e279. 13. Sakone R, Phytinen J, Paakko E, et al. Brain atrophy evaluated by computed tomography in independent and institutionalized hip fracture patients. Disabil Rehabil. 2005 Sep-Oct;27(18e19):1197e1202. 14. Luppa M, Luck T, Weyerer S, et al. Prediction of institutionalization in the elderly. A systematic review. Age Ageing. 2010 Jan;39(1):31e38. 15. Cohen M, Tate J, Baumbusch J. An Uncertain Future for Seniors: BC’s Restructuring of Home and Community Health Care, 2001e2008. Vancouver: Canadian Centre for Policy Alternatives d BC Office. Retrieved at, http://www.policyalternatives.ca/sites/default/files/uploads/publications/ BC_Office_Pubs/bc_2009/CCPA_bc_uncertain_future_full.pdf; 2009. 16. Putnam M, Tang F. Long-term care planning and preparation among persons with multiple sclerosis. Home Health Care Serv Q. 2008;27(2):143e165.
-
(2015)
-
7
17. Multiple Sclerosis Society of Canada. Continuum of Care: Meeting the Needs of Canadians with Multiple Sclerosis. Available from: http://mssociety.ca/en/pdf/socact_homecare-pospaper-feb08-EN.pdf; October 23, 2007. Accessed 31.05.13. 18. Brandis M, Stacom R. Long-term care in the home for people with multiple sclerosis. Care Manag J. 2009;10(3):128e137. 19. Pozzilli C, Brunetti M, Amicosante AM, et al. Home based management in multiple sclerosis: results of a randomised controlled trial. J Neurol Neurosurg Psychiatry. 2002 Sep;73(3):250e255. 20. Canadian Institute for Health Information. National Health Expenditure Trends, 1975e2010. Ottawa: the Institute; 2010. 21. Hader WJ. Disability and survival of multiple sclerosis in Saskatoon, Saskatchewan. Can J Neurol Sci. 2010 Jan;37(1):28e35. 22. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983 Nov;33(11):1444e1452. 23. Allison RS, Millar JHD. Prevalence and familial incidence of disseminated sclerosis: a report to the Northern Ireland Hospitals Authority on the results of a three year survey. Ulster Med. 1954;23(suppl 2): 1e92. 24. Schumacher GA, Beebe G, Kibler RF, et al. Problems of experimental trials of therapy in multiple sclerosis: report by the Panel on Evaluation of Experimental Trials of Therapy in Multiple Sclerosis. Ann N Y Acad Sci. 1965 Mar;122:552e568. 25. Poser CM, Paty DW, Scheinberg LC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol. 1983 Mar;13(3):227e231. 26. McDonald WI, Compston A, Edan G, et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the Diagnosis of Multiple Sclerosis. Ann Neurol. 2001 Jul;50(1): 121e127. 27. Deloire M, Ruet A, Hamel D, et al. Early cognitive impairment in multiple sclerosis predicts disability outcome several years later. Mult Scler. 2010 May;16(5):581e587. 28. Caro J, Ward A, Ishak K, et al. To what degree does cognitive impairment in Alzheimer’s disease predict dependence of patients on caregivers? BMC Neurol. 2002 Aug;2:6. 29. Confavreux C, Vukusic S, Moreau T, Adeleine P. Relapses and progression of disability in multiple sclerosis. N Engl J Med. 2000 Nov;343(20):1430e1438. 30. Leray E, Yaouanq J, Le Page E, et al. Evidence for a two-stage disability progression in multiple sclerosis. Brain. 2010 Jul;133(7): 1900e1913. 31. Lachs M, Bachman R, Williams CS, et al. Violent crime victimization increases the risk of nursing home placement in older adults. Gerontologist. 2006 Oct;46(5):583e589. 32. Russell DW, Cutrona CE, de la Mora A, Wallace RB. Loneliness and nursing home admission among rural older adults. Psychol Aging. 1997 Dec;12(4):574e589. 33. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011 Feb;69(2):292e302. 34. Canadian Institute for Health Information. Canadian Multiple Sclerosis Monitoring System (CMSMS) Metadata. http://www.cihi.ca/ cihi-ext-portal/internet/en/document/typesþofþcare/specializedþserv ices/multipleþsclerosis/cmsms_metadata; Accessed 16.01.14.
Disability and Health Journal
-
(2015)
-
www.disabilityandhealthjnl.com
Research Paper
Predictors of institutionalization for people with multiple sclerosis Lilian U. Thorpe, M.D., Ph.D.a,b,*, Katherine Knox, M.D.c, Rochelle Jalbert, B.Sc.d, June Hyun-Ja Lim, Ph.D.a, Darren Nickel, Ph.D.c, and Walter J. Hader, M.D., F.R.C.P.c a
Department of Community Health and Epidemiology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada b Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan, Canada c Department of Physical Medicine and Rehabilitation, University of Saskatchewan, Saskatoon, Saskatchewan, Canada d College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
Abstract Background: Multiple Sclerosis (MS) is a chronic, progressive disease of the central nervous system with a high prevalence in Canada. While the disease course is highly variable, a significant portion of people with MS may spend more than 10 years living with severe disability, and many of those will eventually require full time institutional care. Despite the high personal and economic cost of this care, little is known about predictors of institutionalization. Objective: The objective of this study was to identify predictors of institutionalization. Methods: Longitudinal data from a university MS clinic database were extracted to explore nursing home placement over time of an urban subgroup. Cox regression analysis was performed with age of MS onset and sex, as well as baseline information obtained at the first MS clinic assessment: MS course, Kurtzke Expanded Disability Status Scale score, and functional system scores. Results: Older age of onset ( p 5 .019) and higher baseline scores in specific functional systems (cerebellar, bowel/bladder, brainstem, and cerebral/mental) were significant ( p 5 .000, p 5 .000, p 5 .001, p 5 .000 respectively) predictors of nursing home placement. Conclusions: Patients with older age of MS onset and those with baseline impairment in specific functional systems (cerebellar, bowel/ bladder, brainstem, and cerebral/mental) may be at higher risk for future institutionalization and should be assessed with particular care to determine potential avenues of support to minimize this. Ó 2015 Elsevier Inc. All rights reserved. Keywords: Multiple sclerosis; Institutionalization; Nursing facilities; Prognosis
Multiple sclerosis (MS) is a chronic degenerative, inflammatory disease of the central nervous system with onset of first symptoms typically in early adulthood. MS survival rates may be increasing, leaving persons with MS living well into older age,1e3 often with significant disability. There is considerable variability in the literature on disability outcomes, with median time to needing a cane between 15 and 32 years after disease onset.4 A relapsingremitting MS (RRMS) course at onset is most common,
Financial disclosure: The authors have no conflicts of interest to declare. This research was supported by a University of Saskatchewan Deans summer student award (project number: Med-12-25). Presentation of abstract at meeting: A poster ‘‘Predictors of Institutionalization for People with Multiple Sclerosis’’ has been accepted for presentation by one of the authors (RJ) at the Fifth Cooperative Meeting of the Consortium of Multiple Sclerosis Centers (CMSC) and Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) in Dallas, Texas from May 28e31, 2014. * Corresponding author. Department of Community Health and Epidemiology, University of Saskatchewan, 107 Wiggins Road, Saskatoon, Saskatchewan, Canada S7N 5E5. Tel.: þ1 3069667977. E-mail address: [email protected] (L.U. Thorpe). 1936-6574/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2014.10.002
offering a more favorable initial prognosis compared to the 10% presenting with a primary progressive MS (PPMS) course at onset. However, 50% of those with RRMS convert to a secondary progressive MS (SPMS) course an average of 15 years after first MS symptoms.5 Progress has been made in the treatment of RRMS, although the impact on long term disability outcomes remains a subject of controversy and no effective treatments are yet available for progressive forms of the disease.6 More information is needed on predicting and planning the care needs of the aging MS population. It has been estimated that over 93,500 people in Canada are living with MS.7 There is considerable geographic variation in prevalence rates across the country.8 Saskatoon, within the province of Saskatchewan, has been reported by Hader & Yee to have a particularly high prevalence rate of 293.8 per 100,000 with a yearly incidence of about 9.5/ 100,000 per year.9 This high prevalence, coupled with the natural features of the disease predict the need for high levels of personal and nursing support related to MS care. Unfortunately, despite the progressive and variable nature of the disease course, there is little literature exploring risk
ARTICLE IN PRESS 2
L.U. Thorpe et al. / Disability and Health Journal
factors for institutionalization specific to MS. Data from the US National Multiple Sclerosis Society (cited by Buchanan, Wang & Ju, 2002) suggested that 20e25% of people with MS eventually end up in long term care, and that the risk might be particularly high in those with bowel dysfunction, poorer health and greater functional decline, as well as in those with greater caregiver burden.10 Studies based on other populations have found that bowel system dysfunction11,12 and cerebellar and cortical system atrophy13 may be increased in institutionalized individuals. A review of general community-based studies14 suggests that age, low self-rated health, functional and cognitive impairment, dementia, prior nursing home placement, and high number of prescriptions are risk factors for nursing home placement. Institutionalization is likely to depend not only on individual needs, but also on the availability and proximity of resources. Saskatchewan is a province large in geography but small in population, which results in people being dispersed over large distances, often removed from the major sources of formal support. This might be expected to result in higher institutionalization rates in general (and for MS patients in particular), and indeed, Saskatchewan was reported to have the second highest ratio of longterm care beds per population in Canada in 2008 at 112.8 beds per 1000 population aged 75þ.15 Another factor potentially contributing to institutionalization is insufficient planning for future care needs. Putnam & Tang suggested that most people with MS are not adequately informed about the availability of services for possible future needs, and frequently do not plan adequately for long term care.16 Even if sufficient alternate community-based supports are available, a lack of readiness to discuss services or knowledge about services could lead to early caregiver burnout and precipitous institutionalization. The Multiple Sclerosis Society of Canada states that individuals themselves generally ‘‘prefer to live in the comfort of their own homes.’’17 In spite of that, little information is available concerning the prevalence of institutionalization or predictors for institutionalization in general or specific to the Canadian MS population. Institutionalization removes people from their familiar environment and family supports, and may decrease their overall autonomy and quality of life. Social isolation within the nursing home is a particular issue for those with MS as they are usually younger and more mentally alert than the general, usually elderly population.18 Health may also be adversely impacted by institutionalization, as home-based care (compared to institutional care) has been associated with improvements in various health dimensions.19 Lastly, institutionalization is very costly to the health care system, with hospitals and other institutions taking up the largest portion of the nation’s expenditures.20 For all of these reasons, it is essential to improve the understanding of the prevalence of and contributors to
-
(2015)
-
institutionalization. While the natural history of MS and risk factors for disease progression in treated and untreated populations have been evaluated,6 these studies generally focus on mobility (i.e., needing a cane to walk), relapse rates, imaging or more recently, quality of life outcomes. Predictors related to institutionalization outcomes identified at clinical assessments might help focus resources optimally to minimize or delay institutionalization. The objective of this study was to identify predictors of institutionalization, using the longitudinal data from a well-established longitudinal MS clinical database maintained at the University of Saskatchewan. We chose to focus on the functional system sub-scores of the EDSS as these sub-scores provide a graded rating of impairment and related function.
Methods The clinical database was initially established as a registry by one of the authors (WH) in 1969 at which time records of patients with MS were extensively searched. Sources included the Saskatoon MS clinic (located in one of the acute care hospitals in Saskatoon), other physicians, nursing homes or home care, and the three city hospitals. Patients were also identified through contact with the MS society and through other provincial records.9 This database was updated every five years under the guidance of WH until 1990 as well as in 1986, 1996 and 2003. Clinical cases coming through the MS clinic at any time or through any of the three Saskatoon hospitals between 2001 and 2005 were added or had their information updated. Yearly phone calls ascertained residence type and living status. The entire group was most recently updated in December 2011 and the records of patients living in Saskatoon at the last contact were additionally updated in summer 2012. Publicly published obituaries were regularly searched by the clinic administrator to update the registry when directly obtained data were missing. The Saskatoon MS Clinic registry as of August 30, 2012 included 1217 patients, of whom 150 had been followed since 1977.21 Standard diagnostic information in the MS registry included clinically-definite, probable, and suspected MS; sex; place of birth; date of birth; place of onset of MS; age at diagnosis of MS; ethnic origin; family history; and date of death. Most records also included disability levels according to Expanded Disability Status Scale (EDSS) scores, functional system subscales,22 and course of MS (relapsing-remitting, secondary progressive, and primary progressive) at assessment times. The MS Clinic registry had additional information available including marital status, education, religion, physical comorbidities, measures of mental health, specific symptoms such as bladder problems, seizure history, events preceding diagnosis, smoking history and income status. Data chosen for our analysis were those with good validity (for example, standardized
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
measurements), consistency and completeness (information gathered consistently over the course of patient follow up period). The protocol for the study was developed collaboratively with the research and clinical team, and ethics approval was obtained from the University of Saskatchewan Biomedical Research Ethics Board. Analyses were conducted on deidentified data. All persons listed in the Saskatoon MS Clinic registry who had an individual, initial assessment at the Saskatoon MS Clinic for the first time prior to 2010, confirming clinically-definite MS by the diagnostic criteria applicable at the time,23e26 and who resided in the Saskatoon Health Region at last contact were included in the initial data selection. The 2010 McDonald diagnostic criteria were not applicable as they were published in 2011, after the cutoff date for study inclusion. When information was incomplete, this was augmented from the clinical record under the supervision of clinicians involved with the project. Records were left in the analysis if there were baseline EDSS and functional systems data and if patients were not already in a nursing home when they were first assessed at the MS clinic. Information collected for this analysis included date of birth, sex, date of onset of first MS symptoms (MS onset), date of initial clinic assessment, course of MS, most recent living situation (own home, own apartment or condominium, assisted living, personal care home, nursing home, other institution or other), first date of permanent admission to a nursing home, date of death and disability level. We use the term nursing home in our document to refer to residential facilities providing 24-h, supervised institutional long-term care services to meet the needs of individuals, usually having heavy needs that cannot appropriately be met in the community. This type of care is formally termed a special-care home by our Ministry of Health. Course of MS included relapsing-remitting, primary progressive, secondary progressive and benign MS. This latter course is most frequently described as minimal disability ten years after onset. After prolonged follow up however, the majority of those with relapsing-remitting course convert to a secondary progressive MS course. MS course at the first clinic assessment was utilized for the analysis as this was determined to be the earliest and most reliable course classification available from the registry. Disability was measured with the Expanded Disability Status Scale (EDSS), the most commonly utilized scale rating neurological impairment in MS. The EDSS is an ordinal scale ranging from a score of zero (no disability) to ten (death due to MS). The EDSS is especially influenced by mobility restrictions and is also calculated according to the subscores from the functional systems: pyramidal, cerebellar, brainstem, sensory, bowel/bladder, visual and cerebral/ mental. Scores for the named functional systems range from zero (normal) to maximum scores of five (cerebellar, brain stem, cerebral/mental) or six (pyramidal, sensory,
-
(2015)
-
3
bowel/bladder, visual), with maximum scores indicating severe impairments in each. Cerebral impairment includes mood alteration only, memory loss, or dementia. Brainstem impairments may include nystagmus, dysarthria, and dysphagia. Cerebellar impairments include coordination deficits. SPSS version 21 was used to perform the data analysis. Data were explored, charted and tabularized in a descriptive format to assess suitability for further statistical analysis. Based on extensive exploration of data distribution and association with outcome variables, cerebral functional system scores (range 0e3) were recoded into a new categorical variable (<2 5 low and score 3 5 high) and brainstem functional system scores (range 0e4) were recoded into a new categorical variable (<2 5 low and scores 3e4 5 high). Available and reliable clinical and demographic variables were entered into a univariate Cox regression model to assess their effects and magnitude of impact on the outcome. The start date for the Cox regression was taken to be the date of the first MS clinic assessment, as baseline assessments with potential predictive value were reliably available from this date. The date of onset of first MS symptoms was generally only estimated by patients, which would result in recall bias. The end date was taken to be the date of first permanent (not respite) admission to a nursing home. Patients were censored if they were not in a nursing home at the last contact. Variables were then added into a multivariate Cox regression model and removed sequentially if their impact was not at a significance of p ! .05. Sex and age of MS onset were left in the model regardless of significance in accordance with established procedures.
Results Of the 1217 patients listed in the Saskatoon MS Clinic registry who had at least one individual assessment at the Saskatoon MS Clinic, 817 had been assessed in the clinic for the first time before January 1, 2010. Of these, 700 resided in the Saskatoon Health Region at the last contact, and 401 of those had both a baseline EDSS score and were not already residing in a nursing home at baseline. Demographics of the study sample are shown in Table 1. A total of 122 males and 279 females were included in this study. Consistent with selection criteria, the most recently enrolled participant was first assessed at the clinic in 2009. Mean follow-up time (years between the first clinic assessment and the last contact) was 13.2 (SD 7.9) years. The mean age at first assessment was 41.8 years (SD 11.6). Males seen in clinic were slightly older at the first clinic assessment than females [mean age 43.3 (SD 11.3) compared to 41.2 (SD 11.6)], but this difference was not statistically significant ( p 5 .85) using parametric statistics for the comparison of means (both male and female distributions were normally distributed in terms of age). The
ARTICLE IN PRESS 4
L.U. Thorpe et al. / Disability and Health Journal
Table 1 Demographics of the study population Male (%) Female (%) Mean age at MS onset (SD) Mean age at first visit (SD) Mean follow up in years (SD) Course of MS at first clinic visit Benign Primary progressive Progressive relapsing Relapsing remitting Secondary progressive Unknown or missing
279 122 31.9 41.8 13.2 N 12 53 7 207 120 2
(69.6) (30.4) (10.7) (11.6) (7.9) (%) (3.1) (13.2) (1.7) (51.6) (29.9) (0.5)
mean age of onset of MS symptoms was 31.9 (SD 10.7) years. Males also had a slightly older reported age of MS onset [mean age 33.3 (SD 10.6) compared to 31.2 (SD 10.7)], which was again not significant ( p 5 .79). At the first clinic visit, 12 patients were diagnosed with benign MS, 206 were diagnosed with relapsing-remitting MS, 120 were diagnosed with secondary progressive MS, 7 were diagnosed with progressive relapsing MS, 53 were diagnosed with primary progressive MS, and 3 had missing diagnoses. Disability scores (EDSS) at the first clinic visit were not significantly different between males and females using the ManneWhitney test for non-normally distributed variables ( p 5 .590). Median EDSS and functional system scores (with ranges in brackets) stratified by MS course are shown in Table 2. In general, patients with primary progressive and secondary progressive MS had the greatest deficits at the first clinic assessment. Fifty-six out a total of 401 (14.0%) were residing in a nursing home at the last contact. The greatest proportion (262/401, or 65.3%) still resided in their own home at the last contact. When stratified by MS type at first assessment we found that the greatest percentage living in a nursing home at the last assessment had been diagnosed with primary progressive MS at the first assessment. People with progressive relapsing MS at baseline had the youngest mean age at initial nursing home admission. However, the only significant predictor of younger age of nursing home admission was younger age of onset of MS. Table 3 shows the results of the multivariate Cox regression analysis for risk of nursing home admission. The final model included older age of MS onset and higher functional system scores at first clinic visit in cerebellar symptoms, bowel/bladder symptoms, brainstem symptoms and
-
(2015)
-
cerebral/mental symptoms as significant predictors of nursing home placement. The impact of gender and other variables initially in the model were not statistically significant. The impact of the significant predictors is illustrated graphically in Figs. 1, 2, and 3. In practical terms, every ten years older at initial assessment increased the hazard of nursing home placement by 1.4 times. Each increase of 1 in the score of baseline cerebellar symptoms increased the hazard of nursing home placement by 1.4 times. Scoring 3 compared to !3 in baseline cerebral-mental symptoms increased the hazard of nursing home placement by 8.7 times. Scoring 3e4 compared to !3 in baseline brainstem symptoms increased the hazard of nursing home placement by 4.7 times. An increase of 1 in the score of baseline bowel-bladder symptoms increased the hazard of nursing home placement by 1.4 times.
Discussion We found that 14% of patients seen in our clinic were admitted to a nursing home over a mean follow up period of just over 14 years (note that this rate is not the final institutionalization rate, for which we would have to follow each patient to death). Institutionalization rates for our participants may, however, actually be lower than rates found elsewhere for a variety of reasons including study participant selection factors, improved community or family supports, low availability of LTC beds in our health region, registry clinic selection bias, and possible non-random factors in data completeness of MS clinic patients. Our study also excluded persons with MS who were institutionalized at first assessment and therefore was likely to find a lower rate than that reported by cross-sectional study designs. Of future interest is a decrease in the LTC bed ratio over time (Peters, R, Saskatchewan Ministry of Health, personal communication, July 23, 2014) and an increasing acuity of patients in LTC over time (Ripley V, Saskatoon Health Region, personal communication, July 16, 2014), and it is unclear how this trend may affect the institutionalization of people with MS. Mental dysfunction was overwhelmingly the strongest predictor for institutionalization even though there were no persons with mental dysfunction in the severe range in our study. This finding is consistent with previous literature in suggesting that early cognitive impairment in MS is
Table 2 Median (range) EDSS and functional system scores by MS course at the first clinic visit MS course at first clinic visit EDSS Pyramidal Cerebellar Brainstem
Sensory
Bowel/bladder
Visual
Cerebral
Scale range Benign Primary progressive Progressive relapsing Relapsing remitting Secondary progressive
0e6 0.50 0.00 1.00 0.00 1.50
0e6 0.00 1.00 0.00 0.00 1.00
0e6 0.00 0.00 0.00 0.00 0.00
0e5 0.00 1.00 0.00 0.00 0.00
0e10 1.25 (0e4) 6.00 (0e9) 2.00 (0e4) 2.00 (0e9) 5.50 (0e9)
0e6 0.00 3.00 0.00 1.00 3.00
(0e1) (3e5) (0e3) (0e6) (0e6)
0e5 0.00 1.00 0.00 0.00 0.00
(0e1) (0e5) (0e2) (0e3) (0e5)
0e5 0.00 0.00 0.00 0.00 0.00
(0e2) (0e4) (0e1) (0e3) (0e4)
(0e3) (0e5) (0e4) (0e5) (0e5)
(0e1) (0e6) (0e6) (0e6) (0e6)
(0e1) (0e6) (0e5) (0e5) (0e6)
(0e1) (0e3) (0e2) (0e3) (0e3)
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
-
(2015)
-
5
Table 3 Cox regression analysis for nursing home placement by variables measured at first clinic visit Hazard ratio 95% CI p value Sex (reference: male) Age at onset of MS per decade Cerebellar system score at first clinic visit Cerebral-mental system score (3 vs. <2) at first clinic visit Brainstem system score (3e4 vs. <2) at first clinic visit Bowel-bladder system score at first clinic visit
1.078 1.379
0.591e1.966 1.055e1.80
0.806* 0.019
1.476
1.205e1.809
0.000
8.731
3.553e21.454
0.000
4.744
1.891e11.903
0.001
1.364
1.186e1.569
0.000
*Not significant but left in the model because of the potential significance of gender.
Fig. 2. Cox regression analysis for nursing home placement; impact of brainstem system score at first clinic visit.
predictive of later increased disability.27 While we do not know the degree of cognitive dysfunction among our participants at latest follow up, in the Alzheimer’s disease literature more severe cognitive impairment has been found to predict the loss of independent functioning. In our study, overall disability as measured by the EDSS and the functional system scores related to vision, other sensory and pyramidal involvement was not predictive of institutionalization. This differential impact of impairments related to specific systems may be explained by their differing impact on independent functioning. Some impairments may be mediated to a greater extent by the use of adaptive technologies. For example, a person with markedly restricted strength in their lower limbs (severe pyramidal system involvement) may manage their activities of daily living independently at a wheelchair level in their own accessible home. In comparison, a severely ataxic individual (severe cerebellar system involvement) who is unable to transfer without assistance, an individual who
has lost bladder or bowel function, or a person frequently aspirating (severe brainstem system involvement) may each have greater challenges in maintaining a safe, independent living environment. Future work should consider the functional impact of the system impairments in MS in predicting longer term independence. Enhanced care, services and technology may reduce the functional impact of these impairments in order to allow more people to stay in their own home. Older age of onset was also a significant predictor of future institutionalization; consistent with the literature that older age is associated with a shorter time interval to disability milestones.28 We did not find that the specific course of MS at first clinic assessment was predictive of institutionalization. This might be explained by the fact that well over half our sample included progressive forms of MS. At our baseline assessment time point many patients had already experienced substantial disability progression.
Fig. 1. Cox regression analysis for nursing home placement; impact of cerebral-mental system score at first clinic visit.
Fig. 3. Cox regression analysis for nursing home placement; impact of age at onset of MS symptoms.
ARTICLE IN PRESS 6
L.U. Thorpe et al. / Disability and Health Journal
Natural history studies have shown that disability progression rates are similar between relapsing and progressive MS after walking is restricted (EDSS 4).29 More recent work also demonstrates that disability rates are similar irrespective of initial MS course once moderate disability in even one functional system has developed (ESDD 3).30 Given our sample demographics and the natural history of later disability progression in MS, it is not surprising that we did not find a statistically significant differential impact of MS course on institutionalization in our analysis. Further research including patients assessed at earlier time points in their disease course might better clarify the differential impacts of various factors, including disease course on institutionalization. This information could provide practical benefits in terms of future planning for patients and their families.
Limitations The most significant limitation of our study involved incomplete information concerning various factors known to be relevant to institutionalization. For example, social support, socioeconomic status, and caregiver burden are important variables which were not consistently collected and therefore not included in our analysis. We were also not able to include individual changes over time in type of living situation (other than confirmation of nursing home placement) because of data over-writing and inconsistencies in the definitions of living situations. Since community residential supports as well as previous institutionalization and admissions to hospital may also be predictive of future long term nursing home placement,31,32 this would be important to capture in the future. We were also unable to include medical and other comorbidities in our analysis as they were not reliably collected in standardized format at the first clinic assessment. This was certainly a limitation for our analysis as one would assume (based on community analyses cited earlier14) that people who moved to nursing homes during our study period were more likely than those who remained in the community to have had comorbid medical and other problems. Another limitation in our study was change in diagnostic patterns over time, related to progress in imaging technology and changes in standard diagnostic criteria. For example, recent revisions to the McDonald criteria33 are thought to enable earlier diagnosis of MS, and changes like this would likely have had an impact on times to eventual institutionalization. We also did not have accurate information on the disease course at MS onset, and had to rely on information obtained at the first clinic assessment. The earlier diagnosis of MS over time could have led to potential cohort changes over time in this longitudinal study. There are also other factors which could have led to cohort changes over time including changes in the referral pattern to the clinic, and the impact of disease-modifying therapies
-
(2015)
-
which were not publically funded in the province of Saskatchewan until 1997. By the first clinic assessment, a significant proportion was already following a secondary progressive MS course, which decreased the utility of survival analysis. The later first presentation to clinic may be in part explained by a referral bias since the Saskatoon MS clinic is based within the Department of Physical Medicine and Rehabilitation. A referral bias to the rehabilitation-based MS clinic may also have impacted institutionalization outcomes. While persons seeking consultation at a rehabilitation-based MS clinic may have more advanced disease, they may also be more actively seeking resources to plan for the future effectively in order to delay or avoid institutionalization. Finally, our results are limited to those residing in the Saskatoon Health Region and those leaving the Saskatoon Health Region over the course of the follow up were not included. This is an important limitation especially since access to community home care supports and long term care access are variable across the province. Future work should include people living initially in rural settings where supports and services for independent living may be different than in urban centers.
Conclusion The final Cox regression analysis found that older age of MS onset and deficits in specific functional systems (cerebellar, bowel/bladder, brainstem, and cerebral/mental) at first clinic assessment were significant predictors of nursing home placement. Particularly important was the contribution of the cerebral/mental functional system score to institutionalization. This suggests that MS patients with these features should be assessed with particular care to determine potential avenues of support to maintain their ability to live independently. In addition to considering institutionalization as an important outcome, there is a need to prospectively collect standardized information about potentially predictive factors related to institutionalization in MS, such as medical comorbidities, self-rated health, prescriptions and psychosocial determinants. On a national level, this may be more feasible in the future through the establishment of a Canadian MS Monitoring System by the Canadian Institute for Health information.34 A more complete understanding of the relative contributions of a broader spectrum of risk factors related to institutionalization should help in developing strategies to keep people with MS in their own homes for longer.
Acknowledgments This study was supported by the College of Medicine, 2012 Dean’s Summer Student Research Projects, University of Saskatchewan. Research resources and space were
ARTICLE IN PRESS L.U. Thorpe et al. / Disability and Health Journal
provided by the Department of Physical Medicine and Rehabilitation, Saskatoon Health Region.
References 1. Bronnum-Hansen H, Koch-Henriksen N, Stenager E. Trends in survival and cause of death in Danish patients with multiple sclerosis. Brain. 2004 Apr;127(Pt 4):844e850. 2. Grytten Torkildsen N, Lie SA, Aarseth JH, et al. Survival and cause of death in multiple sclerosis: results from a 50-year follow-up in Western Norway. Mult Scler. 2008 Nov;14(9):1191e1198. 3. Ragonese P, Aridon P, Salemi G, et al. Mortality in multiple sclerosis: a review. Eur J Neurol. 2008 Feb;15(2):123e127. 4. Tremlett H, Zhao Y, Rieckmann P, Hutchinson M. New perspectives in the natural history of multiple sclerosis. Neurology. 2010 Jun;74(24): 2004e2015. 5. Pittock SJ, Mayr WT, McClelland RL, et al. Disability profile of MS did not change over 10 years in a population-based prevalence cohort. Neurology. 2004 Feb;62(4):601e606. 6. Freedman MS, Selchen D, Arnold DL, et al. Treatment optimization in MS: Canadian MS Working Group updated recommendations. Can J Neurol Sci. 2013 May;40(3):307e323. 7. Statistics Canada. Table 105-1300-Neurological conditions, by age group and sex, household population aged 0 and over, 2010/2011, occasional (number unless otherwise noted), CANSIM (database). http:// www5.statcan.gc.ca/cansim/a26?lang5eng&retrLang5eng&id51051 300&paSer5&pattern5&stByVal51&p151&p2531&tabMode5 dataTable&csid5; Updated September 19, 2013. Accessed 19.12.13. 8. Poppe AY, Wolfson C, Zhu B. Prevalence of multiple sclerosis in Canada: a systematic review. Can J Neurol Sci. 2008 Nov;35(5):593e601. 9. Hader WJ, Yee IM. Incidence and prevalence of multiple sclerosis in Saskatoon, Saskatchewan. Neurology. 2007 Sep 18;69(12): 1224e1229. 10. Buchanan RJ, Wang S, Ju H. Analyses of the minimum data set: comparisons of nursing home residents with multiple sclerosis to other nursing home residents. Mult Scler. 2002 Dec;8(6):512e522. 11. AlAmeel T, Andrew MK, MacKnight C. The association of fecal incontinence with institutionalization and mortality in older adults. Am J Gastroenerol. 2010 Aug;105(8):1830e1834. 12. Carton H, Loos R, Pacolet J, et al. A quantitative study of unpaid caregiving in multiple sclerosis. Mult Scler. 2000 Aug;6(4):274e279. 13. Sakone R, Phytinen J, Paakko E, et al. Brain atrophy evaluated by computed tomography in independent and institutionalized hip fracture patients. Disabil Rehabil. 2005 Sep-Oct;27(18e19):1197e1202. 14. Luppa M, Luck T, Weyerer S, et al. Prediction of institutionalization in the elderly. A systematic review. Age Ageing. 2010 Jan;39(1):31e38. 15. Cohen M, Tate J, Baumbusch J. An Uncertain Future for Seniors: BC’s Restructuring of Home and Community Health Care, 2001e2008. Vancouver: Canadian Centre for Policy Alternatives d BC Office. Retrieved at, http://www.policyalternatives.ca/sites/default/files/uploads/publications/ BC_Office_Pubs/bc_2009/CCPA_bc_uncertain_future_full.pdf; 2009. 16. Putnam M, Tang F. Long-term care planning and preparation among persons with multiple sclerosis. Home Health Care Serv Q. 2008;27(2):143e165.
-
(2015)
-
7
17. Multiple Sclerosis Society of Canada. Continuum of Care: Meeting the Needs of Canadians with Multiple Sclerosis. Available from: http://mssociety.ca/en/pdf/socact_homecare-pospaper-feb08-EN.pdf; October 23, 2007. Accessed 31.05.13. 18. Brandis M, Stacom R. Long-term care in the home for people with multiple sclerosis. Care Manag J. 2009;10(3):128e137. 19. Pozzilli C, Brunetti M, Amicosante AM, et al. Home based management in multiple sclerosis: results of a randomised controlled trial. J Neurol Neurosurg Psychiatry. 2002 Sep;73(3):250e255. 20. Canadian Institute for Health Information. National Health Expenditure Trends, 1975e2010. Ottawa: the Institute; 2010. 21. Hader WJ. Disability and survival of multiple sclerosis in Saskatoon, Saskatchewan. Can J Neurol Sci. 2010 Jan;37(1):28e35. 22. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983 Nov;33(11):1444e1452. 23. Allison RS, Millar JHD. Prevalence and familial incidence of disseminated sclerosis: a report to the Northern Ireland Hospitals Authority on the results of a three year survey. Ulster Med. 1954;23(suppl 2): 1e92. 24. Schumacher GA, Beebe G, Kibler RF, et al. Problems of experimental trials of therapy in multiple sclerosis: report by the Panel on Evaluation of Experimental Trials of Therapy in Multiple Sclerosis. Ann N Y Acad Sci. 1965 Mar;122:552e568. 25. Poser CM, Paty DW, Scheinberg LC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol. 1983 Mar;13(3):227e231. 26. McDonald WI, Compston A, Edan G, et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the Diagnosis of Multiple Sclerosis. Ann Neurol. 2001 Jul;50(1): 121e127. 27. Deloire M, Ruet A, Hamel D, et al. Early cognitive impairment in multiple sclerosis predicts disability outcome several years later. Mult Scler. 2010 May;16(5):581e587. 28. Caro J, Ward A, Ishak K, et al. To what degree does cognitive impairment in Alzheimer’s disease predict dependence of patients on caregivers? BMC Neurol. 2002 Aug;2:6. 29. Confavreux C, Vukusic S, Moreau T, Adeleine P. Relapses and progression of disability in multiple sclerosis. N Engl J Med. 2000 Nov;343(20):1430e1438. 30. Leray E, Yaouanq J, Le Page E, et al. Evidence for a two-stage disability progression in multiple sclerosis. Brain. 2010 Jul;133(7): 1900e1913. 31. Lachs M, Bachman R, Williams CS, et al. Violent crime victimization increases the risk of nursing home placement in older adults. Gerontologist. 2006 Oct;46(5):583e589. 32. Russell DW, Cutrona CE, de la Mora A, Wallace RB. Loneliness and nursing home admission among rural older adults. Psychol Aging. 1997 Dec;12(4):574e589. 33. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011 Feb;69(2):292e302. 34. Canadian Institute for Health Information. Canadian Multiple Sclerosis Monitoring System (CMSMS) Metadata. http://www.cihi.ca/ cihi-ext-portal/internet/en/document/typesþofþcare/specializedþserv ices/multipleþsclerosis/cmsms_metadata; Accessed 16.01.14.