Effects of health screening for adults with intellectual disability: a pooled analysis

INTRODUCTION

People with intellectual disability have much unrecognised disease and inadequate health promotion.1–3 Barriers to providing satisfactory health care include difficulties gaining access, short consultation times, and communication difficulties.4 Beneficial effects of health screening in adults with intellectual disability have been reported;5–7 however, recruitment in this population is difficult,4,8 and individual studies have been small and effect estimates unstable. This study has addressed this by conducting a pooled analysis to assess the overall benefits of health screening for adults with intellectual disability.

METHOD

The Medline, PsycINFO, EMBASE, and CINAHL databases were searched to identify studies from articles published between 1980 and 2009. The terms: ‘intellectual disability’, ‘trial OR study OR control’, ‘adults’, and ‘short health screening OR health assessment’ were used. To be included, the study population had to be adults with intellectual disability, include a prospective intervention with a comparison group, and have allocation occurring at GP level or higher, to guard against carry-over effects within general practice. Two randomised trials were found,6,7 and one cohort study.5 One study investigating the effect of repeated screens was excluded due to allocation occurring within general practice.9

Original data were obtained from each included study. Post-intervention unadjusted 1-year cumulative incidence was calculated for all outcomes. A multi-level mixed-effects logistic regression was used to compute the odds ratios and 95% confidence intervals of achieving a health-related event in the health-screening group compared to the usual care group, with adjustment for clustering and study.

RESULTS

Three studies were identified that met the research criteria. All study participants lived in the community.

• Study 1: Cooper et al conducted a matched cohort study in 2002–2003 in Scotland.5 Participants who received the intervention were individually matched on age, sex, and ability with adults from the neighbouring locality. The health screen used was The C21 Health Check.10 Nurses extracted data from GP notes, then interviewed and examined participants before reviewing the findings with a GP to formulate recommendations.

• Study 2: Lennox et al conducted a cluster randomised controlled trial in 2000–2001 in Australia.6 Randomisation clusters were groups of participants who shared a residence or GP. The health screen used was the Comprehensive Health Assessment Program (CHAP). Carers completed a health history, before the adult with intellectual disability visited their GP, who reviewed the history, performed an examination, and completed an action plan.

• Study 3: Lennox et al conducted a second trial (2 × 2 factorial) in 2004–2005.7 Randomisation clusters were groups of participants who shared a GP. Interventions were a CHAP health screen and a health diary. There was no interaction between interventions, so health outcomes among those screened were compared with those receiving usual care, ignoring whether they received a diary.

In all studies, clinical outcomes were collected from medical records at 12 months post-intervention. In studies 2 and 3, data were extracted using the same variable definitions. Where these were compatible to those used in study 1, data were pooled. A disease diagnosis was defined as new if it had not been noted previously in the GPs' records or specialists' letters. In studies 2 and 3, data extractors were masked; in study 1 they were not.

The pooled analysis provided data on 407 adults assigned to receive health screening and 388 who received usual care. Fifty-eight per cent of participants were male, 69% had mild or moderate intellectual disability, and the median age was 37 years (range 19–79 years). Overall, a greater proportion of males than females received health screening (55% versus 47%); there was no difference in age or level of disability.

The intervention group generally received far more sensory testing and provision of health-promotion or disease-prevention activities (Table 1). Notably, there were increases in vision (odds ratio [OR] 4.2; 95% confidence interval [CI] = 2.3 to 7.4) and hearing (OR 10.8; 95% CI = 3.4 to 34.3) tests performed and problems identified, as well as in key immunisations, obesity recording, and women's health screens. Typically, odds ratios varied by study, with wide and overlapping confidence intervals. For activities embedded in day-to-day practice, differences were less marked. Overall clinical activity and disease detection increased with screening, although neither notably nor significantly for most individual diseases (Table 2). No adverse effects from use of the screening tool were reported.

DISCUSSION