We followed the PRISMA guidelines for systematic review in this review [16].
Literature search
The first author conducted a systematic literature search in February 2018 in the following databases: Scopus, Embase, CINAHL, and MEDLINE. The first author devised the search strategy in collaboration with a research librarian. The search string was adjusted to each database, using the MeSH, Emtree, or exact major subject heading (MM)/exact subject heading (MH) terms for each database. The following search strings were used:
(Comorbidity [mesh, emtree, MM] OR Multimorbidity [emtree] OR (Multimorbid* or multi-morbid* or comorbid* or co-morbid* or multidisease* or multi-disease* or frail* or vulnerab*) OR ((multipl* or cooccur*or co-exist* or co-occur*) adj3 (ill* or disease* or condition* or syndrom* or disorder* or symptom* or medication* or health*))) AND (Aged [mesh, emtree, MM/MH] OR Aged, 80 and over [mesh, MM] OR Senescence [emtree] OR Very elderly [emtree] OR (old* or elder* or aged or geriatic* or senescence or senior* or senium or centenarian* or nonagenarian* or octogenarian* or genrontolog* or "late life")) AND (Patient Satisfaction [mesh, emtree, MM] OR Patient attitude [emtree, MM] OR Patient preference [emtree] OR ((Patient*) adj3 (Attitude* or opinion* understanding* or perspective* or satisfaction* or preference* or view* or standpoint* or perception* or experience*))) AND (Continuity of Patient Care[Mesh, MM/MH] OR Patient care [emtree] OR Delivery of Health Care, Integrated [mesh, MM] OR Integrated Health Care Systems [emtree] OR Intersectoral Collaboration [mesh, emtree] OR Cooperative Behavior [mesh] OR Cooperation [emtree] OR "contin* of patient care" OR "Patient Care Contin*" OR "Contin* of Care" OR "Care Contin*" OR "Coordinat* care" OR "multidisciplinary care" OR "multi-disciplinary care" OR "Intersectoral Collaboration*" OR "Intersectoral Cooperation*" OR "Integrated Delivery System*" OR "Cooperative Behavior*" OR "Compliant Behavior*").
Study selection
After the literature search, the first author screened all studies twice using the titles and abstracts, with a gap of a week between each screening, using the systematic review management program Covidence [17]. Last author was consulted when in doubt. Next, both authors conducted full-text screenings separately. In case they disagreed on certain studies to be included in the full-text screening, first, these two authors discussed such studies and if they did not arrive at an agreement, then the other authors on the research team read and discussed the articles to arrive at a consensus.
Inclusion criteria
We included studies that fulfilled the following inclusion criteria:
- The studies had to explore the patients’ viewpoints and address aspects such as their experience, and opinions about transitions, their relationship with their health care providers and relatives, the transition of information, and consistency of care and personnel and the health care system they were navigating within. The study participants had to be patients aged 65 years or older.
- The study participants had to be patients with multimorbidity. If this fact was not explicitly stated, only studies with participants aged 85 years or older were included owing to the high proportion of multimorbidity in this age group [4].
We excluded studies that were:
- not written in English, Danish, Norwegian, or Swedish
- not peer-reviewed, published articles.
Data extraction
Analysis
We included only one quantitative study and therefore we chose to code and analyzed it integrated with the other studies, which were qualitative. Therefore, all results are presented narratively. We analyzed all studies using thematic synthesis [18]. The synthesis consisted of three stages. In the first stage, we coded each article line by line according to its meaning and content about the manner in which older patients with multimorbidity experienced transitions, their relationship with their health care providers and relatives, the transition of information, and consistency of care and personnel, creating several initial codes. In the second stage, we formed descriptive themes to capture the meaning of the initial codes. In the first and second stages, we stayed close to the original findings in the included studies. Creating the descriptive themes offered us the possibility to go beyond the content in the original studies. In stage three, we created analytical themes, which means we used the descriptive themes to find answers to our research questions [18]. We used QSR International’s NVivo11 qualitative data analysis software to conduct the synthesis [19].
Quality appraisal
We used the CoreQ checklist for assessment of study quality [20]. In rating the quality of the included study, we focused only on domain 2, Study design (items 9, 10, 11, 12, 14, 15, 16, and 17), and domain 3, Analysis and findings (items 26, 29, and 30). Each study was assigned one point if the items were identified; therefore, a study could obtain 0–11 points, with 11 being the highest score, thus indicating highest quality. We rated the quality high if the score was 8 to 11, medium if 4 to7 and low if 0 to 3. We only included articles scoring 4 or higher. The first author conducted the quality assessment, and last author reviewed the results.