Articles were retrieved from 5 databases. Although exact terms varied by database, the same general process was followed in each. 2 searches were conducted in each database: 1 for items on home telehealth in management of our target chronic diseases (Concepts 1 and 2; see below), and 1 for items on the economic aspects of home telehealth (Concepts 1 and 3; see below).
Concept 1: Home telehealth. Exact terms varied by database. MeSH terms used in the Medline search were [exp telemedicine], [exp telenursing], [exp telemetry], [exp remote consultation], [exp telecommunications], and [exp computer communication networks]. The closest equivalents to MeSH terms were used when selecting EMBASE subject headings. Text words were fairly expansive and included terms such [telemonitor* or tele-monitor*], [telecare or tele-care], and [telehome or tele-home].
Concept 2: Chronic diseases. Exact terms varied by database, but typically consisted of the chronic disease names and their variants. MeSH terms used in the Medline search were [exp chronic disease], [exp diabetes mellitus], [exp asthma], [exp pulmonary disease, chronic obstructive], [exp arthritis], [exp cardiovascular diseases], and [exp kidney diseases].
Concept 3: Economics. Exact terms varied by database. MeSH terms used in the Medline search included [economics], [exp “cost and cost analysis”], [“cost control”], [“cost savings”], [“cost allocation”], [cost-benefit analysis], [“cost of illness”], [“cost sharing”], [“deductibles and coinsurance”], [medical savings accounts], [exp health care costs], [direct service costs], [drug costs], [employer health costs], [hospital costs], [health expenditures], [capital expenditures], [“value of life”], [exp economics, hospital], [exp economics, medical], [economics, nursing], [economics, pharmaceutical], [exp “fees and charges”] and [exp budgets].
The full list of terms used in each database is excluded here for reasons of space, but can be provided by the authors upon request. Yields were as follows:
# of items retrieved (pre-screening)
|Web of Science||12,688|
Note that the extent of de-duplication within each database was dependent on database functionality. The percentage of unique items in the yields listed above therefore varies.
 Arthritis articles were later excluded.