Table 1

Search results: 13 papers evaluating the fit note

Study	Paper	Year of data collection	Recruitment and sampling	Data collection	Funder	Study design and method	Demographic variation	FNs	Study population			Outcomes§
Study	Paper	Year of data collection	Recruitment and sampling	Data collection	Funder	Study design and method	Demographic variation	FNs	Patients/ employees	GPs and FFW	Employer	1	2	3	4
1	Chenery 25	January to June 2012	Adults selected from the Labour Force Survey (LFS). The Fit Note Survey was attached to the last of five LFS interviews.	Survey—computer assisted telephone interviews	Department for Work and Pensions	Cross-sectional Logistic regression	Age, gender, SES, health condition, size of organisation, sector worked in, self-report health, disability status, health condition		1398 employees			Y		Y
2	Coole et al 26	June 2011 to December 2011	Selected from a service evaluation	Fit note comments	National Institute for Health Research	Cross-sectional Content analysis		1212		1 practice (712 FNs) 1 FFW (500 FNs)		Y
3	Hussey et al 27	4 years before (April 2006 to March 2010) and 3 years (July 2010 to June 2013)	Selected all available data from the THOR-GP surveillance scheme	Information reported by GPs with training in occupational health to diploma level†	UK Health and Safety Executive (THOR contract number: HSE JN4243)	Longitudinal Multilevel random effects Poisson regression model fitted to the monthly data count	Type of health condition		5517 patients (835 patients since introduction of the fit note)	250 GPs		Y	Y
4	Coole *et al* 29	November 2013 to May 2014	A total of 272 GP practices were invited to participate in the study.	Data from fit note copies and questionnaires	Institute of Occupational Safety and Health	Cross-sectional Quantitative data were analysed descriptively. Free-text comments using thematic content analysis	Type of health condition	94		11 GPs		Y	Y
4+5	Coole et al 28	Study 4+5 combined	Study 4+5 combined	Study 4+5 combined			Type of health condition	94+498		11 GPs	13 employer organisations	Y	Y	Y
5	Coole et al 30	May 2013 to March 2014	A combination of opportunistic and random sampling	Postal questionnaires				498				Y	Y	Y
6	Shiels *et al* 31	2001–2002 + October 2011 to January 2013	Practices were invited to take part from 5 areas of UK. The second project involved additional data from an evaluation of 19 practices sited in 3 FFWS pilot sites.	Carbonised fit note pads	Department for Work and Pensions	Longitudinal Logistic regression: multilevel mixed effects model	Patient (age, gender, type of health condition, social deprivation), GP (sex age partner/locum status, full time/part time) GP practice (size, location, GP deprivation)	58 700	25 000 patients	49 practices		Y	Y		Y
	Shiels et al 32	Late 2011 to January 2013				Longitudinal and cross-sectional Multivariate logistic regression models		79 815	33 768	68 practices including 3 FFWS pilot sites
	Gabbay et al 33	2001–2002 + late 2011 to January 2013						25 061	10 984			Y	Y
	Gabbay et al 34‡	2001–2002 + late 2011 to January 2013						8074	3288			Y	Y		Y
	Shiels et al 35	Late 2011 to January 2013						38 934 episodes	31 453			Y Y	Y
	Gabbay et al 36‡	Late 2011 to January 2013						8127	3361			Y			Y
7	Digital NHS 37	December 2014 to March 2017	207 CCGs across the UK, 70 had more than 95% coverage, 25 had less than 5%.	NHS electronic fit note data	NHS Digital	Descriptive analysis	Type of health condition, location	5 603 986*				Y	Y		Y

Bold denotes grey literature.
*62.1% of all fit notes prescribed in England during this period to people aged 18–65.
†Patients with CMD.
‡Patients with depression only.
§Outcome 1: maybe fit for work outcome. Outcome 2: work solutions. Outcome 3: return to work outcomes. Outcome 4: length of sickness absence.
CMD, common mental disorder; FFWS, Fit For Work scheme; FN, fit note; GP, general practitioner; NHS, National Health Service; SES, socioeconomic status; THOR, The Health and Occupational Research Network; CCG, clinical commisioning groups.