sjl_sc: Compute Jankowski's MCTQ sleep-corrected social jetlag

In ropensci/mctq: Munich ChronoType Questionnaire Tools

Compute Jankowski's MCTQ sleep-corrected social jetlag

Description

sjl_sc() computes the Jankowski's (2017) sleep-corrected social jetlag for standard, micro, and shift versions of the Munich ChronoType Questionnaire (MCTQ).

sjl_sc_rel() is just a wrapper for sjl_sc() with abs = FALSE.

Please note that the Jankowski (2017) did not proposed a "relative" sleep-corrected social jetlag, but the user may consider using it.

Usage

sjl_sc(so_w, se_w, so_f, se_f, abs = TRUE, method = "shorter")

sjl_sc_rel(so_w, se_w, so_f, se_f, method = "shorter")

Arguments

so_w	An hms object corresponding to the local time of sleep onset on workdays from a standard, micro, or shift version of the MCTQ questionnaire. You can use so() to compute it for the standard or shift version.
se_w	An hms object corresponding to the local time of sleep end on workdays from a standard, micro, or shift version of the MCTQ questionnaire.
so_f	An hms object corresponding to the local time of sleep onset on work-free days from a standard, micro, or shift version of the MCTQ questionnaire. You can use so() to compute it for the standard or shift version.
se_f	An hms object corresponding to the local time of sleep end on work-free days from a standard, micro, or shift version of the MCTQ questionnaire.
abs	(optional) a logical object indicating if the function must return an absolute value (default: TRUE).
method	(optional) a string indicating which method the function must use to compute the social jetlag. See the Methods section to learn more (default: "shorter").

Details

Standard MCTQ functions were created following the guidelines in Roenneberg, Wirz-Justice, & Merrow (2003), Roenneberg, Allebrandt, Merrow, & Vetter (2012), and from The Worldwide Experimental Platform (theWeP, n.d.).

\muMCTQ functions were created following the guidelines in Ghotbi et al. (2020), in addition to the guidelines used for the standard MCTQ.

MCTQ^{Shift} functions were created following the guidelines in Juda, Vetter, & Roenneberg (2013), in addition to the guidelines used for the standard MCTQ.

See the References section to learn more.

Class requirements

The mctq package works with a set of object classes specially created to hold time values. These classes can be found in the lubridate and hms packages. Please refer to those package documentations to learn more about them.

Rounding and fractional time

Some operations may produce an output with fractional time (e.g., "19538.3828571429s (~5.43 hours)", 01:15:44.505). If you want, you can round it with mctq:::round_time().

Our recommendation is to avoid rounding, but, if you do, make sure that you only round your values after all computations are done. That way you avoid round-off errors.

Value

• If abs = TRUE, a Duration object corresponding to the absolute sleep-corrected social jetlag.

• If abs = FALSE, a Duration object corresponding to the relative sleep-corrected social jetlag.

The output may also vary depending on the method used.

Guidelines

In an article published in 2017, Konrad S. Jankowski argued that the original formula for computing the social jetlag (SJL) captures not only the misalignment between social and biological time, but also the sleep debt resulting from sleep deprivation during workdays. Jankowski then proposed the following guideline for a sleep-corrected social jetlag (SJL_{sc}) computation.

Notes

• The Jankowski's alternative is disputed. We recommend seeing Roenneberg, Pilz, Zerbini, & Winnebeck (2019) discussion about it (see item 3.4.2).

• For MCTQ^{Shift}, the computation below must be applied to each shift section of the questionnaire.

• Due to time arithmetic issues, sjl_sc() does a slightly different computation by default than those proposed by the author mentioned above. See vignette("sjl-computation", package = "mctq") for more details.

• If you are visualizing this documentation in plain text, you may have some trouble understanding the equations. You can see this documentation on the package website.

For standard and micro versions of the MCTQ

\textrm{If } SD_{W} > SD_{F} \; \& \; SE_{W} \leq SE_{F} \; , \; SJL_{sc} = | SE_{F} - SE_{W} |

\textrm{Else } \; , \; SJL_{sc} = | SO_{F} - SO_{W} |

Where:

• SJL_{sc} = Jankowski's sleep-corrected social jetlag.

• SO_{W} = Local time of sleep onset on workdays.

• SE_{W} = Local time of sleep end on workdays.

• SO_{F} = Local time of sleep onset on work-free days.

• SE_{F} = Local time of sleep end on work-free days.

* W = Workdays; F = Work-free days.

For the shift version of the MCTQ

\textrm{If } SD_W^{M/E/N} > SD_F^{M/E/N} \; \& \; SE_W^{M/E/N} \leq SE_F^{M/E/N} \; , \; SJL_{sc}^{M/E/N} = | SE_F^{M/E/N} - SE_W^{M/E/N} |

\textrm{Else } \; , \; | SJL_{sc}^{M/E/N} = SO_F^{M/E/N} - SO_W^{M/E/N} |

Where:

• SJL_{sc}^{M/E/N} = Jankowski's sleep-corrected social jetlag in a particular shift.

• SO_W^{M/E/N} = Local time of sleep onset between two days in a particular shift.

• SE_W^{M/E/N} = Local time of sleep end between two days in a particular shift.

• SO_F^{M/E/N} = Local time of sleep onset between two free days after a particular shift.

• SE_F^{M/E/N} = Local time of sleep end between two free days after a particular shift.

* W = Workdays; F = Work-free days, M = Morning shift; E = Evening shift; N = Night shift.

Methods for computing the sleep-corrected social jetlag

There are different approaches to compute the sleep-corrected social jetlag (SJL_{sc}). By default, sjl_sc() uses an approach that we call "the shorter interval approach" ("shorter").

The topics below provide a simple explanation of each method supported by sjl_sc(). To get a detail understating of this methods, see vignette("sjl-computation", package = "mctq").

• "difference"

By using method = "difference", sjl_sc() will do the exact computation proposed by Jankowski, i.e., SJL_{sc} will be computed as the linear difference between SO_f/SE_f and SO_W/SE_W (see the Guidelines section).

We do not recommend using this method, as it has many limitations.

• "shorter"

This is the default method for sjl_sc(). It's based on the shorter interval between SO_f/SE_f and SO_W/SE_W, solving most of the issues relating to SJL_{sc} computation.

• "longer"

The "longer" method uses the same logic of the "shorter" method, but, instead of using the shorter interval between SO_f/SE_f and SO_W/SE_W, it uses the longer interval between the two, considering a two-day window.

This method may help in special contexts, like when dealing with shift-workers that have a greater than 12 hours distance between their sleep hours.

References

Ghotbi, N., Pilz, L. K., Winnebeck, E. C., Vetter, C., Zerbini, G., Lenssen, D., Frighetto, G., Salamanca, M., Costa, R., Montagnese, S., & Roenneberg, T. (2020). The \muMCTQ: an ultra-short version of the Munich ChronoType Questionnaire. Journal of Biological Rhythms, 35(1), 98-110. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1177/0748730419886986")}

Jankowski K. S. (2017). Social jet lag: sleep-corrected formula. Chronobiology International, 34(4), 531-535. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1080/07420528.2017.1299162")}

Juda, M., Vetter, C., & Roenneberg, T. (2013). The Munich ChronoType Questionnaire for shift-workers (MCTQ^{Shift}). Journal of Biological Rhythms, 28(2), 130-140. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1177/0748730412475041")}

Roenneberg T., Allebrandt K. V., Merrow M., & Vetter C. (2012). Social jetlag and obesity. Current Biology, 22(10), 939-43. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.cub.2012.03.038")}

Roenneberg, T., Pilz, L. K., Zerbini, G., & Winnebeck, E. C. (2019). Chronotype and social jetlag: a (self-) critical review. Biology, 8(3), 54. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.3390/biology8030054")}

Roenneberg, T., Wirz-Justice, A., & Merrow, M. (2003). Life between clocks: daily temporal patterns of human chronotypes. Journal of Biological Rhythms, 18(1), 80-90. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1177/0748730402239679")}

The Worldwide Experimental Platform (n.d.). MCTQ. https://www.thewep.org/documentations/mctq/

See Also

Other MCTQ functions: fd(), gu(), le_week(), msf_sc(), msl(), napd(), sd24(), sd_overall(), sd_week(), sdu(), sjl(), sjl_weighted(), so(), tbt()

Examples

## Scalar example

so_w <- hms::parse_hm("02:00")
se_w <- hms::parse_hm("10:00")
so_f <- hms::parse_hm("01:00")
se_f <- hms::parse_hm("08:00")

sjl_sc(so_w, se_w, so_f, se_f)
#> [1] "3600s (~1 hours)" # Expected
sjl_sc(so_w, se_w, so_f, se_f, abs = FALSE)
#> [1] "-3600s (~-1 hours)" # Expected (negative sjl_sc)
sjl_sc_rel(so_w, se_w, so_f, se_f) # Wrapper function
#> [1] "-3600s (~-1 hours)" # Expected (negative sjl_sc)
sjl(msl(so_w, sdu(so_w, se_w)), msl(so_f, sdu(so_f, se_f)))
#> [1] "5400s (~1.5 hours)" # Expected

so_w <- hms::parse_hm("22:00")
se_w <- hms::parse_hm("06:00")
so_f <- hms::parse_hm("01:00")
se_f <- hms::parse_hm("06:00") # sd_w > sd_f & se_w <= se_f

sjl_sc(so_w, se_w, so_f, se_f) # sjl_sc = | se_f - se_w |
#> [1] "0s" # Expected
sjl_sc(so_w, se_w, so_f, se_f, abs = FALSE)
#> [1] "0s" # Expected
sjl_sc_rel(so_w, se_w, so_f, se_f) # Wrapper function
#> [1] "0s" # Expected
sjl(msl(so_w, sdu(so_w, se_w)), msl(so_f, sdu(so_f, se_f)))
#> [1] "5400s (~1.5 hours)" # Expected

so_f <- hms::as_hms(NA)

sjl_sc(so_w, se_w, so_f, se_f)
#> [1] NA # Expected

## Vector example

so_w <- c(hms::parse_hm("00:00"), hms::parse_hm("01:00"))
se_w <- c(hms::parse_hm("08:00"), hms::parse_hm("07:00"))
so_f <- c(hms::parse_hm("01:00"), hms::parse_hm("01:00"))
se_f <- c(hms::parse_hm("09:00"), hms::parse_hm("09:00"))

sjl_sc(so_w, se_w, so_f, se_f)
#> [1] "3600s (~1 hours)" "0s" # Expected
sjl_sc(so_w, se_w, so_f, se_f, abs = FALSE)
#> [1] "3600s (~1 hours)" "0s" # Expected
sjl_sc_rel(so_w, se_w, so_f, se_f) # Wrapper function
#> [1] "3600s (~1 hours)" "0s" # Expected
sjl(msl(so_w, sdu(so_w, se_w)), msl(so_f, sdu(so_f, se_f)))
#> [1] "3600s (~1 hours)" "3600s (~1 hours)" # Expected

## See other examples in '?sjl()'

ropensci/mctq documentation built on Dec. 15, 2024, 1:12 p.m.