Susan M Smith, Emma Wallace, Tom O'Dowd, Martin Fortin

Many people with chronic disease have more than one chronic condition, which is referred to as multimorbidity. The term comorbidity is also used but this is now taken to mean that there is a defined index condition with other linked conditions, for example diabetes and cardiovascular disease. It is also used when there are combinations of defined conditions that commonly co-exist, for example diabetes and depression. While this is not a new phenomenon, there is greater recognition of its impact and the importance of improving outcomes for individuals affected. Research in the area to date has focused mainly on descriptive epidemiology and impact assessment. There has been limited exploration of the effectiveness of interventions to improve outcomes for people with multimorbidity. To determine the effectiveness of health-service or patient-oriented interventions designed to improve outcomes in people with multimorbidity in primary care and community settings. Multimorbidity was defined as two or more chronic conditions in the same individual. We searched MEDLINE, EMBASE, CINAHL and seven other databases to 28 September 2015. We also searched grey literature and consulted experts in the field for completed or ongoing studies. Two review authors independently screened and selected studies for inclusion. We considered randomised controlled trials (RCTs), non-randomised clinical trials (NRCTs), controlled before-after studies (CBAs), and interrupted time series analyses (ITS) evaluating interventions to improve outcomes for people with multimorbidity in primary care and community settings. Multimorbidity was defined as two or more chronic conditions in the same individual. This includes studies where participants can have combinations of any condition or have combinations of pre-specified common conditions (comorbidity), for example, hypertension and cardiovascular disease. The comparison was usual care as delivered in that setting. Two review authors independently extracted data from the included studies, evaluated study quality, and judged the certainty of the evidence using the GRADE approach. We conducted a meta-analysis of the results where possible and carried out a narrative synthesis for the remainder of the results. We present the results in a 'Summary of findings' table and tabular format to show effect sizes across all outcome types. We identified 18 RCTs examining a range of complex interventions for people with multimorbidity. Nine studies focused on defined comorbid conditions with an emphasis on depression, diabetes and cardiovascular disease. The remaining studies focused on multimorbidity, generally in older people. In 12 studies, the predominant intervention element was a change to the organisation of care delivery, usually through case management or enhanced multidisciplinary team work. In six studies, the interventions were predominantly patient-oriented, for example, educational or self-management support-type interventions delivered directly to participants. Overall our confidence in the results regarding the effectiveness of interventions ranged from low to high certainty. There was little or no difference in clinical outcomes (based on moderate certainty evidence). Mental health outcomes improved (based on high certainty evidence) and there were modest reductions in mean depression scores for the comorbidity studies that targeted participants with depression (standardized mean difference (SMD) -2.23, 95% confidence interval (CI) -2.52 to -1.95). There was probably a small improvement in patient-reported outcomes (moderate certainty evidence) although two studies that specifically targeted functional difficulties in participants had positive effects on functional outcomes with one of these studies also reporting a reduction in mortality at four year follow-up (Int 6%, Con 13%, absolute difference 7%). The intervention may make little or no difference to health service use (low certainty evidence), may slightly improve medication adherence (low certainty evidence), probably slightly improves patient-related health behaviours (moderate certainty evidence), and probably improves provider behaviour in terms of prescribing behaviour and quality of care (moderate certainty evidence). Cost data were limited. This review identifies the emerging evidence to support policy for the management of people with multimorbidity and common comorbidities in primary care and community settings. There are remaining uncertainties about the effectiveness of interventions for people with multimorbidity in general due to the relatively small number of RCTs conducted in this area to date, with mixed findings overall. It is possible that the findings may change with the inclusion of large ongoing well-organised trials in future updates. The results suggest an improvement in health outcomes if interventions can be targeted at risk factors such as depression, or specific functional difficulties in people with multimorbidity.