. 2018 Oct 18;13(10):e0205741. doi: 10.1371/journal.pone.0205741

Table 2. Codesheet from the included articles.

Study	HRV	Effect direction (+ = positive, - = negative)	Significant at the p < .05 level? (y/n)	Time HR assessment	Stress Measure	Study design; Year; Country	Sample Description	N analysis; N reported	Age Mean; Range; %Female; Stress division	Remarks
(Bishop et al., 2003)[65]	HR_w	+	y	every 30 min during a workday	JDC	C; -; Singapore	Singapore Police Officers	108; 118	26.9; 19–50; -; SD
(Borchini et al., 2014)[40] and (R. Borchini, personal communication, March 1st, 2017)	HF_l	-	n	2 24 hour days continuous	JDC and ERI both used to identify high strain	L; 2010; Italy	CVD susceptible nurses	36	38.1; -; 83.3; E/R D/C
	HF_wln	-	y
	PNN50_l	-	n
	PNN50_wln	-	n
	RMSSD_lln	-	n
	RMSSD_wln	-	n
	SDNN_lln	-	n
	SDNN_wln	-	y
	HR_lln	+	n
	HR_wln	+	n
(Butterbaugh et al., 2003)[66]	HR_r	ne	n	-	JDC	C; -; -	Newly employed female nurses	58	-; -; 100; D/C
(Chandola et al., 2008)[67]	HF_r	-	y	5 min RR was used	JDC	L; 1985–2004; Great Britain	Civil servants	3290	-; 35–56; -; Mdn
(Chandola et al., 2008)[67]	SDNN_r	-	y	5 min RR was used	JDC	L; 1985–2004; Great Britain	Civil servants	3290	-; 35–56; -; Mdn
(Clays et al., 2011)[43]	HF_wln	-	n	24 hours including workday, HRV measures are based on 24 h	JDC	C; 1976–1978; Belgium	Healthy male factory workers	653; 770	47; 40–55; 0; Sum
	PNN50_wln	-	n
	SDNN_wln	-	n
	HR_wln	+	y
(Collins et al., 2005)[68]	HF_w(l)ln	-	y	48 hours, including work and rest days	JDC	C; -; United States	Healthy employed day shift working men from a community health plan and N = 6 from a stress reduction program	34; 36	45; 35–59; 0; Tri
	HF_w	-	n
	SDNN_w(l)ln	+	n
	SDNN_w	-	n
	HR_w(l)ln	+	n
	HR_w	+	y
(van Doornen et al., 2009)[69]	HF_l	-	n	24 hour workday	Burnout (Maslach)	C; -; Netherlands	Male managers of a Dutch telecommunications company	88	43.3; -; 0; HLC
	HF_n	-	n
	HF_w	-	n
	HR_l	-	n
	HR_n	+	n
	HR_w	+	n
(Ekstedt et al., 2004)[70]	HR_n	+	y	24 hour, but HR measured at rest before awaking at 7 am +/- 1 hour	Burnout (Shirom Melamed)	C; -; Sweden	Employees of IT company	24	30.5; 24–43; 58.3; HLC	The significant effect must be interpreted with caution, it is part of a multiple regression analysis and burnout group is entered as a dummy.
(Eller et al., 2011)[38]	HF_r	-	n	18 hour ECG starting on a workday in both 2006 and 2008, but only 15 min logtransformed seated rest in analysis	ERI	L; 2006–2008; Denmark	White collar workers in the public administration males	61	51.2; -; 0; E/R
	HF_r	-	y
	HR_r	+	y
	HR_r	+	y
(Eriksson et al., 2016)[71]	HR_r	+	y	5–10 min resting period	JDC	C; 2001–2004; Sweden	Working population	1552	46; 24–71; 52; D/C
(Fauvel et al., 2001)[72]	HR_r	ne	n	Measured during 15 minutes of seated rest	JDC	C; 1995–2001*; France	workers of a chemical company	281	37.3;18–55; 8; HL	Top 20% was considered a high strain group
	HR_t	+	n
	HF_r	+	n
	HF_t	+	n
(Hamer et al., 2006)[45]	HR_r	+	n	10 min BP (measured last 5 min of a 10 min resting period)	ERI	C; 2003–2004; Great Britain	full time employed men	92	33.1; -; 0; E/R
(Hamer et al., 2006)[45]	HR_t	-	y	8 min BP measured during a role playing and mirror tracing tasks	ERI	C; 2003–2004; Great Britain	full time employed men	92	33.1; -; 0; E/R
(Hanson et al., 2001)[59]	HF_w(l)	-	n	During a working day, but for the office clerks the measurements continued into the evening (until 21.30)	ERI	C; -; Netherlands	Health professionals and office clerks	70	36.3; -; 44; E/R	Only seated periods were analyzed
(Henning et al., 2014)[47]	RMSSD_n	ne	n	24 hour Amb measurement, but the unit of analysis is data between 2 am and 4 am	Burnout (Copenhagen Burnout Inventory)	L; -; New Zealand	junior doctors	17	-; 20-?; 65; Mean	In the conclusion it states that there were no doctors with burnout, so there is actually nothing to compare
(Henning et al., 2014)[47]	RMSSD_wl	ne	n		Burnout (Copenhagen Burnout Inventory)	L; -; New Zealand	junior doctors	17	-; 20-?; 65; Mean
(Hernández-Gaytan et al., 2013) [73]	HF_w	-	n	24 hour ECG workday, 8 hour shift and 16 hour on call time	JDC	C; 2007–2008; Mexico	resident doctors	54	-; 23–36; 33; Mdn
(Hernández-Gaytan et al., 2013) [73]	SDNN_w	-	n	24 hour ECG workday, 8 hour shift and 16 hour on call time	JDC	C; 2007–2008; Mexico	resident doctors	54	-; 23–36; 33; Mdn
(Herr et al., 2015)[41] and (R. Herr, personal communication, February 22nd, 2017)	HF_n	+	y	24 hour ECG	OI	C; 2007; Germany	White collar workers	179	46.4; -; 0; Sum	Sum is total OI scale
	HF_wln	+	n
	RMSSD_n	+	y
	RMSSD_wln	+	n
	SDNN_n	+	y
	SDNN_wln	+	n
	HF_n	-	n				Blue collar workers	222	44.3; -; 0; Sum
	HF_wln	-	n
	RMSSD_n	-	n
	RMSSD_wln	-	n
	SDNN_n	-	n
	SDNN_wln	-	n
(Hintsanen et al., 2007)[39]	HR_r	-	n	3 min controlled breathing	ERI	C; 2001–2002; Finland	Employed people working full time males	406	32.2; 24–39; 0; E/R
	HF_r	+	n
	PNN50_r	+	n
	RMSSD_r	+	n
	HR_r	+	n					457	32.3; 24–39; 100; E/R
	HF_r	-	n
	PNN50_r	-	y
	RMSSD_r	-	y
(Jarczok et al., 2016)[61]	RMSSD_n	-	y	24 hour workday continuous	ERI	C; 2010–2012; Germany	Mannheim Industrial Cohort (MICS)	9924; 9937	41.9; 18–65; 19; E/R
(Jarczok et al., 2016)[61]	RMSSD_wl	-	y	24 hour workday continuous	ERI	C; 2010–2012; Germany	Mannheim Industrial Cohort (MICS)	9924; 9937	41.9; 18–65; 19; E/R
(Johnston et al., 2016)[74]	HR_w	+	n	2 12 hour workdays	JDC and ERI	C; -; Great Britain	Qualified nurses in a general hospital on medical and surgical wards	100	36.4; -; 93; Sum
(Jönsson et al., 2015)[3] and (P. Jönsson, personal communication, February 16 th, 2017)	HR_t	-	n	1 hour ECG during task and recovery with a baseline reading	Burnout (Shirom Melamed)	Lab; -; -	Employed population with N = 14 Former ED (Burnout) patients, n = 17 pre ED stage workers and n = 20 controls	51	48.7; 33–61; 51; HLC
	HF_t	+	n	1 hour ECG during task and recovery with a baseline reading	Burnout (Shirom Melamed)	Lab; -; -		51	48.7; 33–61; 51; HLC
(Kang et al., 2004)[49]	HF_r	-	n	5 minutes in the morning	JDC	C; 2003; South-Korea	Male shipyard workers	169	46.7; 41-?; 0; D/C
(Kang et al., 2004)[49]	SDNN_r	-	n	5 minutes in the morning	JDC	C; 2003; South-Korea	Male shipyard workers	169	46.7; 41-?; 0; D/C
(Karhula et al., 2014)[75]	HR_n	-	n	3 non consecutive 24 hour days including a morning shift, night shift and recovery day. Data used for analysis was at least 4h of sleep after which the 30 min segment with the lowest heart rate was used for analysis.	JDC	C; 2008; Finland	Female nurses	95	47.2; 31–59; 100; mdn and mean	mdn and mean (to get a greater contrast)
	HF_n	ne	n
	RMSSD_n	ne	n
(Kotov and Revina, 2012)[50]	HF_w	-	y	8 hour workday	Burnout (Boiko)	C; -; Russia	First-aid doctors	44; 84	-; 26–65; 56; HLC	Both coping strategy groups show a negative effect
	PNN50_w	+	n							The effect is negative for a task-oriented coping strategy. The effect is positive for an emotion oriented strategy.
	RMSSD_w	-	n							The effect is negative for a task-oriented coping strategy. The effect is positive for an emotion oriented strategy.
	SDNN_w	-	n							Burnout (Alarm stage vs no Burnout). Article uses coping strategies as comparator. The effect is negative for a task-oriented coping strategy. The effect is positive for an emotion oriented strategy.
(Lee et al., 2010)[76]	HF_r	+	n	Measured 3 times in each subject after completion of 1 day, 1 night and 1 eveningshift in a 5 min rest period after 5 min of rest.	JDC	C; -; South-Korea	Employees of consumer goods company	56; 140	29.1; 25–44; 0; D/C
(Lennartsson et al., 2016)[52]	HF_r	-	y	5 minutes in the morning in supine position	Burnout (Shilom-Melamed)	Lab; -; Sweden	Employed, working and on sick leave burnout patients, non-clinincal burnout subjects and healthy controls	161	40; 20–65; 60; HLC/HL	The effects are only significant in men.
	RMSSD_r	-	y
	SDNN_r	-	y
(Loerbroks et al., 2010)[30]	RMSSD_l	ne	n	24 hour wln measure	ERI	C; 2003–2004; Germany	Employees from an airplane manufacturer	591; 657	41.6; 17–65; 12; E/R	Some positive, some negative, only age group 35–44 negative significant effect
	RMSSD_l	ne	n		JDC
	RMSSD_n	ne	n		ERI
	RMSSD_n	ne	n		JDC
	RMSSD_w	ne	n		ERI
	RMSSD_w	ne	n		JDC
(Morgan et al., 2002)[32]	HF_r	+	y	10 min in supine position	Burnout (Maslach)	C; -; United States	Soldiers	41	-; -; -; HL
(Moya-Albiol et al., 2010)[48]	HR_w	-/ne	y/n	3 times a day 30 min during seated rest	Burnout (Maslach)	C; -; Spain	Full- time school teachers	64; 80	42.8; -; 80; Mean	This effect was caused by HR at the middle of the day, which was significantly negatively correlated, the beginning of the day was positive, and the end of the day negatively related. Those were non-significant.
(Nomura et al., 2005)[77]	HR_r	+	n	After 5 min of rest measures were taken at rest	JDC	C; 2003; Japan	Employees from IT company	396; 437	30; 24–39; 0; D/C	20% highest were allocated to high job strain group
(Ohira et al., 2011)[42]	HR_r	+	y	During baseline and 2 learning tasks	JDC	Lab; -; Japan	Full time employed men	20	32.6; -; 0; D/C	D/C (but with median split in the sample)
	HF_r	+	n.r.
	HF_t	+	n.r.
(Poorabdian et al., 2013)[64]	HR_?	+	n.r.*	-	JDC	C; 2007–2009; Iran	Male personnel at a petrochemical plant	500	42.5; 22–64; 0; HL	* The authors presented a Chi-square. The percentage of people with the highest heart rate was highest in the high job strain group. (only the omnibus test is presented for all 12 categories)
(Rau, 2001)[44]	HR_n	ne	n.r.	Every 15 minutes over 24 hours during a normal working day	JDC	C; 1985-?; Sweden	Employed hypotensive (n = 74) and hypertensive (n = 75) men	81; 149	50.1; 35–55; 0; Mean	Mean (they use z scores). In the regression analysis, both control and support scales have a negative effect. The demand scale is not significant as it is not reported, the direction is therefore unclear.
(Rau, 2001)[44]	HR_w	ne	n.r.	Every 15 minutes over 24 hours during a normal working day	JDC	C; 1985-?; Sweden	Employed hypotensive (n = 74) and hypertensive (n = 75) men	81; 149	50.1; 35–55; 0; Mean
(Riese et al., 2004)[78]	HR_w(l)ln	ne	n	2 days, for 24 hours on a workday and one on a leisure day	JDC	L; 1997–1998; Netherlands	Healthy female nurses	159	35.9; 25–50; 100; Median	Median with a distinction of the four quadrants each year. High job strain year 1 and 2, yes or no. (results in 4 groups, y-y, n-n, y-n, n-y)
(Riese et al., 2004)[78]	RMSSD_w(l)ln	ne	n	2 days, for 24 hours on a workday and one on a leisure day	JDC	L; 1997–1998; Netherlands	Healthy female nurses	159	35.9; 25–50; 100; Median
(Salavecz et al., 2010)[79]	HF_l	-	y	Measured over the working day	JDC	C; -; Hungary	Women working in Budapest	169	-; -; 100; -	They report on data after work
(Teisala et al., 2014)[80]	RMSSD_w	-	n	3 24 hour workdays, HRV measures are based on 24 h, not all participants three days. One day (n = 10, two days (n = 70), three days (n = 1).	Burnout (bergen)	C; -; -	Employed people	81	34; 26–40; 0; Mean
(Uusitalo et al., 2011)[51]	HF_w	ne	n	2 36 hour workdays	ERI	C; -; Finland	Healthy hospital workers	19; 22	42; 24–57; 95; E/R	On day 2 it was significant, not on day 1
	RMSSD_w	-	y/n
	SDNN_w	ne	n
(van Amelsvoort et al., 2000)[44]	HR_n	-	n	24 hour workday	JDC	C; -; Netherlands	Shift workers and daytime workers as controls, working in the manufacturing industry, waste incinerator industry and hospitals	135; 155	30.8; 18–55; 19; D/C	For SDNN_n the contrast between high D, H control and low stress was significant)
	HR_w	+	y
	HF_n	-	n
	HF_w	-	n
	SDNN_n	-	n
	SDNN_w	+	n
(Vrijkotte et al., 2000)[21]	HR_l	+	y	2 24 hour workdays and 1 24 h non workday	ERI	C; 1996–1997; Netherlands	White collar workers of a computer company	109	47.2; 37–57; 0; E/R
	HR_n	+	n
	HR_w	+	y
	RMSSD_l	-	n
	RMSSD_n	-	n
	RMSSD_w	-	n

JDC = Job demands control; ERI = Effort reward imbalance; OI = Organizational injustice; C = Cross-sectional; L = Longitudinal; Lab = Laboratory; HR = Heart rate; HF = High frequency; RMSSD = Root mean square of successive differences; PNN50 = percentage of adjacent cycles that are greater than 50 ms apart; HRV = Heart rate variability; SDNN = Standard deviation of all N-N intervals; RR = R to R intervals; SD = Standard deviation; E/R = Effort divided by reward; D/C = Demand divided by control; Mdn = Median; Sum = Sumscore; Tri = Triangulation of data; HL = Based on high low scores; HLC = Based on clinical high low scores; _l = measured during leisure time; _r = measured during rest; _w = measured during a workday; wln = measured during a period including workday, leisure time and night;_n = measured during a night; _t = Measured during a task; n.r. = Not reported; ne = No effect.