Meteorological factors and childhood diarrhea in Peru, 2005–2015: a time series analysis of historic associations, with implications for climate change

Table 3 Associations between meteorological factors and incidence rate of childhood clinic visits for diarrhea, Peru, 2005–2015, stratified by piped water access

	Piped water access^a
	Low provinces (N = 64) IRR (95% CI)	High provinces (N = 56) IRR (95% CI)	Transitional provinces (N = 74) IRR (95% CI)
Temperature across 3 weeks prior to diarrhea cases^b	1.017 (1.007, 1.027)	1.043 (1.034, 1.051)	1.042 (1.033, 1.051)
1-week temperature lag^c	1.004 (1.000, 1.009)	1.021 (1.016, 1.026)	1.013 (1.008, 1.019)
2-week temperature lag^c	1.008 (1.003, 1.014)	1.016 (1.011, 1.021)	1.019 (1.014, 1.024)
3-week temperature lag^c	1.004 (0.999, 1.009)	1.005 (1.000, 1.010)	1.009 (1.004, 1.013)
Moderate/strong El Niño period	1.042 (1.008, 1.077)	1.023 (1.000, 1.046)	1.026 (0.996, 1.057)
Dry season	1.036 (1.014, 1.059)	1.007 (0.989, 1.026)	1.012 (0.992, 1.032)
Rotavirus vaccine era (2010–2015)^d	0.955 (0.895, 1.019)	0.892 (0.851, 0.934)	0.901 (0.862, 0.942)
Year (secular trend)^e	0.974 (0.962, 0.985)	0.957 (0.946, 0.968)	0.970 (0.960, 0.981)

IRR incidence rate ratio, CI confidence interval
Multivariable model: IRRs are controlled for other variables in the model/table, and for province.
^a “Low piped water access” provinces were defined as those in which < 60% of households had access to a piped water connection in all study years (2005–2015), or all but 1 year. “High piped water access” provinces were those in which ≥60% of households had access to a piped water connection in all study years, or all but 1 year. “Transitional piped water access” provinces were those that did not fall into either category, i.e., those that transitioned from lower piped water access (< 60% of households with a piped water connection) to higher piped water access (≥60% of households with a piped water connection) between 2005 and 2015.
^b Combined effect of temperature across 3 weeks prior to weekly diarrhea report.
^c The 1-week temperature lag is the effect of temperature in the week before the diarrhea cases, the 2-week lag refers to the week before that, etc.
^d Compared to the pre-rotavirus vaccine era (2005–2009).
^e Continuous term for year.

ISSN: 1476-069X