gms | German Medical Science

Background: ARDS (acute respiratory distress syndrome) results from either pulmonary or extrapulmonary induced damage of lung parenchyma. The most common causes are polytrauma, pneumonia and sepsis. ARDS is a life-threatening condition with in-hospital mortality up to 40%. Although strategies to decrease mortality in ARDS attracted much interest in research, many survivors of ARDS suffer from impaired health-related quality of life (HRQoL) and long term disability. Research is increasingly addressing these important outcomes However, there is still little evidence on the determinants of HRQoL and return to work (RtW) in patients surviving ARDS.

Research Questions: Which are the characteristics (e.g. participants, study designs, follow-up intervals) of published scientific studies investigating determinants of HRQol and/or RtW in survivors of ARDS?

What domains of determinants (socio-demographic, clinical, psycho-social or care related) for HRQol and/or RtW have been investigated in the published scientific literature?

Methods: Observational studies reporting statistical associations between determinants of HRQoL and/or RtW in patients surviving ARDS were considered for this systematic review. No interventional studies were included due to concerns considering their ecological validity. The language of the full-text papers needed to be either English or German. Primary outcome for this systematic review was HRQoL. HRQoL had to be assessed by a self-report instrument (scale/s or single-item measure). Secondary outcome was RtW assessed by either self-report or routine data (e.g. provided by health insurances). The search included electronic databases (PubMed, Embase, PsycInfo, and Web of Science) from inception until August 13, 2014 and references of existing systematic and non-systematic reviews. Assessment of study eligibility and data extraction were conducted by four researchers (working in two pairs) independently.

Results: The electronic search of the databases resulted in 4559 studies (after electronic duplicate check). 82 research reports remained after the first screening on the basis of title and abstract. After closer examination of the full text papers 27 research reports based on 21 study populations met the eligibility criteria. The sizes of the study populations ranged from n=15 to n=168. In the included studies a total number of 1309 participants were examined. The distribution of the different study types was: i) 18 (67%) prospective cohort studies, ii) 5 (19%) retrospective cohort studies and iii) 4 (15%) cross-sectional studies. HRQoL was assessed by five generic and two disease specific self-report instruments. Most frequently (78%) HRQoL was assessed by the generic 36-Item Short-Form Health Survey (SF-36). In the seven studies which investigated determinants of RtW the outcome was operationalized in terms of general return to any work (seven studies) or return to the previous position (one study). HRQoL was analyzed in 26 (96%) and RtW in seven (26%) studies. Therefore only one study solely focused on RtW and neglected HRQoL. There was variability between the studies regarding the number (range from 1-4) and length of follow-ups (range from two weeks after endotracheal extubation to 13 years after ICU admission). Six (22%) of the studies investigated socio-demographic determinants (gender and age). Most determinants (18) identified by this review can be subsumed under clinical and disease related factors (e.g. comorbidities, length of ICU and hospital stay, performance in pulmonary function testing) and 14 (52%) of the studies conducted statistical analyses related to these clinical determinants. The association between neuropsychological domains and HRQol was analyzed in two (7%) studies.

Discussion: This systematic review revealed high degree of heterogeneity in the included studies in important characteristics, such as sample size, study type, outcome measurement instruments, and number and length of follow-ups. The main focus was on clinical and disease related determinants of HRQoL and RtW. Other potentially influencing and modifiable factors relating to e.g. quality of care have been neglected so far.

Implications: Future observational research in survivors of ARDS should use longitudinal study designs and comprehensive, standardized measurements of determinants and outcomes, enabling comparison between studies and evidence synthesis. Research efforts should be geared towards including care related factors in the studies to provide a substantial improvement of medical care for patients with ARDS.