Factors associated with acute care service use after epilepsy hospitalisation in people with intellectual disability

BACKGROUND

This study aimed to identify factors associated with unplanned acute hospital readmission and emergency department (ED) presentation after hospitalisation for epilepsy in people with intellectual disability (ID).

METHODS

This study is a retrospective cohort study using linked administrative datasets. We identified 3293 people with ID aged 5-64 years with a hospitalisation for epilepsy between 2005 and 2014 in New South Wales, Australia. We examined unplanned readmission and ED presentation within 30 or 365 days and associations with demographic, socio-economic and health status variables. Modified Poisson regression with robust estimation was used to model outcomes within 30 days. Negative binomial regression was used to account for the overdispersion of the data and to model 365-day outcome rates.

RESULTS

Around half of the cohort had an unplanned readmission and ED presentation within 365 days of the index hospitalisation. In fully adjusted models, being female, being a young adult and having a longer or acute care index admission, mental and physical comorbidities and a history of incarceration were associated with an elevated risk of readmission or ED presentation. The strongest association was observed between history of self-harm and 365-day readmission (incidence rate ratio 2.15, 95% confidence interval 1.41-3.29).

CONCLUSIONS

Socio-demographic, justice and health factors are associated with unplanned readmission and ED presentation risk after hospitalisation for epilepsy in people with ID. Interventions targeting improving continuity of care should be tailored for individuals and their support workers. The findings also emphasise the importance of person-centred multidisciplinary care across different health sectors.

Reviewing causes of death of individuals with intellectual disability in New South Wales, Australia: a record-linkage study

BACKGROUND

People with intellectual disability (ID) have a much higher mortality rate than the general population. To reduce the rate of mortality of people with ID, it is critical that causes of death are properly understood, recorded and reported. Formal reviews of causes of death are used in some countries to ensure that causes of death are accurate. To date, the impact of these formal reviews on understanding causes of death of people with ID has not been quantified.

METHODS

The study aimed to quantify the impact of formal reviews of deaths on the understanding of causes of death of people with ID who died while living in residential care. Individuals (851) with ID who died in residential care in New South Wales (NSW), Australia, between 1 December 2002 and 31 December 2013, who had a cause of death recorded in both the NSW Cause of Death Unit Record File (COD-URF; cause of death recorded at time of death) and NSW Ombudsman dataset (cause of death recorded after in-depth review) were included in the study. We assessed agreement in coding for cause of death by comparing the International Classification of Diseases 10th Revision (ICD-10) codings at three levels of diagnostic specificity, for both underlying and additional causes of death. We conducted our analysis through both descriptive comparison and through two boosted regression trees.

RESULTS

Approximately half of the underlying causes of death were different after review by the NSW Ombudsman compared with the COD-URF. Certain causes of death (determined by ICD-10 chapter) were less likely to predict matches between the dataset than others, with individuals with mental, behavioural and neurodevelopmental disorders recorded in the COD-URF least likely to have a matching cause of death in NSW Ombudsman dataset. For deaths where there was no agreement at any level between the datasets, a high level of unknown causes of death was recorded.

CONCLUSIONS

Formal review of deaths of people with ID in residential care is important to determining true causes of death and therefore developing appropriate health policy for people with ID.

Using linked administrative data to determine the prevalence of intellectual disability in adult prison in New South Wales, Australia

BACKGROUND

Studies of the representation of people with intellectual disability (ID) in custody report widely inconsistent findings that reflect variation in how ID is defined and the methods employed for identification. Using linked administrative data may be of utility in studies of the representation of people with ID in custody. However, this approach requires an understanding of the purpose of and factors influencing identification in disparate administrative datasets.

METHODS

This study uses linked administrative data encompassing disability, health and corrections data for the year 2014 to estimate the prevalence of ID in adult custody and explore how ID representation within administrative data impacts prevalence estimates and what patterns of identification reveal about support service access for this group.

RESULTS

This study finds that 4.3% of the New South Wales adult custody population had an identified ID. Prisoners with ID were younger, more likely to have had a previous custodial episode and more likely to be Indigenous than the general prison population. Identification of ID across linked administrative datasets is uneven, which, if used in isolation, would result in variation in prevalence estimates according to source data.

CONCLUSIONS

The utilisation of linkage data from a broad range of health and support services including custody offers a comprehensive identification methodology. Inconsistency in the identification of ID across datasets indicates a potential disjuncture between prisoners with ID and support services, which may have relevance for efforts to reduce reincarceration of those in this population.

Survival of children and adolescents with intellectual disability following gastrostomy insertion

BACKGROUND

Positive health outcomes have been observed following gastrostomy insertion in children with intellectual disability, which is being increasingly used at younger ages to improve nutritional intake. This study investigated the effect of gastrostomy insertion on survival of children with severe intellectual disability.

METHODS

We used linked disability and health data of children and adolescents who were born in Western Australia between 1983 and 2009 to compare survival of individuals with severe intellectual disability by exposure to gastrostomy status. For those born in 2000-2009, we employed propensity score matching to adjust for confounding by indication. Effect of gastrostomy insertion on survival was compared by pertinent health and sociodemographic risk factors.

RESULTS

Compared with children born in the 1980s-1990s, probability of survival following first gastrostomy insertion for those born in 2000-2009 was higher (2 years: 94% vs. 83%). Mortality risk was higher in cases than that in their matched controls (hazard ratio 2.9, 95% confidence interval 1.1, 7.3). The relative risk of mortality (gastrostomy vs. non-gastrostomy) may have differed by sex, birthweight and time at first gastrostomy insertion. Respiratory conditions were a common immediate or underlying cause of death among all children, particularly among those undergoing gastrostomy insertion.

CONCLUSIONS

Whilst gastrostomy insertion was associated with lower survival rates than children without gastrostomy, survival improved with time, and gastrostomy afforded some protection for the more vulnerable groups, and earlier use appears beneficial to survival. Specific clinical data that may be used to prioritise the need for gastrostomy insertion may be responsible for the survival differences observed.

Positive spillover effects of prescribing requirements: increased cardiac testing in patients treated with trastuzumab for HER2+ metastatic breast cancer

BACKGROUND

Cardiotoxicity is a concern in patients on trastuzumab therapy, and cardiac function assessment is a recommended practice. In 2006, trastuzumab was publically subsidised for human epidermal growth factor receptor-2 early stage breast cancer with a requirement for cardiac testing prior to and during treatment.

AIM

To investigate the spillover effects of this requirement on testing rates in metastatic patients treated with trastuzumab where no monitoring requirements are applied.

METHODS

We examined cardiac testing (echocardiography or multiple-gated acquisition scan) in 3779 women with metastatic breast cancer receiving trastuzumab between December 2001 and February 2010 and used interrupted time-series analyses to estimate changes in testing rates. The main outcome measures were the proportion of eligible patients, by quarter, receiving a cardiac function test pretreatment and during trastuzumab therapy.

RESULTS

Only 21% of women had a cardiac function test pretreatment, and 47% were tested at some point during the first year of trastuzumab therapy. The introduction of mandatory cardiac testing for early breast cancer was associated with an immediate 8% increase (95% confidence interval, 2-14%) in pretreatment cardiac testing and an immediate 7% increase (95% confidence interval, 4-10%) in testing during therapy in metastatic patients. Testing rates during therapy increased steadily from early 2005, coinciding with the release of interim results from several trastuzumab trials reporting cardiac-safety outcomes.

CONCLUSION

The introduction of mandatory cardiac testing for early stage disease spilled over to the metastatic setting. While deviation from guidelines may be warranted in some cases, this study suggests underutilisation of cardiac testing among patients treated with trastuzumab in the metastatic setting.

Prior antiretroviral therapy experience protects against zidovudine-related anaemia

OBJECTIVE

We investigated the use of antiretroviral therapy regimens containing zidovudine or stavudine, using data from the TREAT Asia HIV Observational Database (TAHOD), a multicentre, prospective, observational study of an HIV-infected cohort in the Asia-Pacific Region.

METHODS

A proportional hazards regression analysis of factors associated with the time to discontinuation of initial regimens containing zidovudine or stavudine and a logistic regression analysis to identify factors associated with a diagnosis of anaemia within 6 months of commencement of zidovudine in initial or subsequent regimens were performed.

RESULTS

Patients who started zidovudine were more likely to stop within the first 9 months of treatment than those who started on stavudine; the reverse was true after 9 months. Anaemia (haemoglobin<or=10 g/dL) occurred in the first 6 months in 57 of 433 patients (13%) on zidovudine. Baseline anaemia was the strongest predictive factor for subsequent anaemia, and prior antiretroviral therapy (ART) experience was protective for development of anaemia.

CONCLUSIONS

These data support baseline haemoglobin testing and avoidance of zidovudine if the patient is anaemic. The protective effect of prior ART for development of anaemia on zidovudine supports the short-term safety of a stavudine to zidovudine switch with routine haemoglobin monitoring in this cohort. Further studies in resource-poor settings of longer term efficacy and toxicities of ART switch strategies are needed.

Predictive factors for immunological and virological endpoints in Thai patients receiving combination antiretroviral treatment

BACKGROUND

Routine CD4 count and HIV viral load monitoring is a financial barrier in developing countries.

METHODS

We assessed factors associated with CD4 counts < or =200 cells/microL and detectable viral load in Thai HIV-infected patients receiving antiretroviral therapy (ART) at the HIV Netherlands Australia Thailand Research Collaboration and the Thai Red Cross AIDS Research Centre (HIV-NAT). Univariate and multivariate Cox proportional hazards models for multiple treatment failures were used to determine factors related to CD4 counts < or =200 cells/microL and detectable viral load. Multivariate Cox proportional hazards models for CD4 counts < or =200 cells/microL were developed with and without viral load in order to build models applicable to contexts in which viral load is not available.

RESULTS

Four hundred and seventeen patients were included in the study. Fifty-four per cent were male, and the median CD4 count and log(10) viral load at baseline were 283 cells/microL and 4.3 log(10) HIV-1 RNA copies/mL, respectively. Independent factors related to CD4 count < or =200 cells/microL were CD4 count at baseline [hazards ratio (HR) 0.20/100 cells/microL; 95% confidence interval (CI) 0.17-0.23] and changes in CD4 count (HR 0.22/100 cells/microL; 95% CI 0.17-0.28). Factors in multivariate models (in which viral load was considered for inclusion) were CD4 count at baseline (HR 0.21/100 cells/microL; 95% CI 0.18-0.24), changes in CD4 count (HR 0.25/100 cells/microL; 95% CI 0.19-0.32) and detectable viral load (HR 1.94; 95% CI 1.20-3.13). Predictive factors (independent of viral load) were triple ART or highly active antiretroviral therapy (HAART) (HR 0.28; 95% CI 0.22-0.36) and detectable viral load at baseline (HR 2.96; 95% CI 2.24-3.91). Conclusions CD4 count at baseline and changes in CD4 count were important in predicting CD4 counts < or =200 cells/microL. Triple ART and detectable viral load at baseline were important in predicting detectable viral load.

The 48-week efficacy of once-daily saquinavir/ritonavir in patients with undetectable viral load after 3 years of antiretroviral therapy

OBJECTIVES

To evaluate the efficacy and safety of once-daily saquinavir-soft-gel-capsules/ritonavir (SQV-SGC/RTV) 1600 mg/100 mg plus dual nucleoside reverse transcriptase inhibitors (NRTIs) in HIV-infected patients with plasma viral load (pVL) <50 HIV-1 RNA copies/mL following 3 years of antiretroviral therapy.

METHODS

A total of 69 patients with pVL <50 copies/mL after 162 weeks of antiretroviral treatment started SQV-SGC/RTV 1600 mg/100 mg once-daily while continuing dual NRTIs. Previous treatment consisted of 66 weeks of treatment with a half/full dose of zidovudine (ZDV)/zalcitabine (ddC), followed by 2 years of SQV-SGC twice a day (bid) plus ZDV/lamivudine (3TC) or didanosine (ddI)/stavudine (d4T). Efficacy (pVL), safety and immunological changes (CD4 cell counts) were evaluated after 48 weeks in this open-label, single-arm prospective study.

RESULTS

SQV-SGC/RTV once-daily was well tolerated. No patient changed regimens or was lost to follow-up. After 48 weeks, 63 of 69 patients (91%) had pVL <50 copies/mL (five of the six remaining patients had pVL <400 copies/mL, and one patient had an unexplained rise to 39 500 copies/mL, which decreased to <50 copies/mL 12 weeks later). Median CD4 count increased from 534 cells/muL at the start of the SQV-SGC/RTV once-daily treatment to 664 cells/muL (P<0.001). Compared to the preceding 48 weeks on bid SQV-SGC, the CD4 cell count improved significantly on once-daily SQV-SGC/RTV (P<0.001).

CONCLUSIONS

These data support the use of SQV-SGC/RTV 1600 mg/100 mg once-daily with two NRTIs as a convenient, safe and cost-saving regimen to maintain viral suppression and CD4 counts for 48 weeks in this preselected cohort on highly active antiretroviral therapy (HAART) with pVL <50 copies/mL. The CD4 count rise may be a result of continued immune reconstitution in patients with well-controlled infection.