A time series analysis of weather variables and all-cause mortality in the Kasena-Nankana Districts of Northern Ghana, 1995–2010

Figures & data

Table 1. Summary statistics of daily mortality data by age and sex in the Kassena-Nankana Districts (KNDs) of Northern Ghana (1995–2010)

Population groups	Daily mortality count(n)	Mean deaths	SD	Min–max	Percent (%)
All	31,144	5.3	2.4	0.1–21.2	100
Age group
0–4	8,551	1.5	1.3	0.0–11.0	27.5
5–19	2,315	0.4	0.5	0.0–4.6	7.4
20–59	9,417	1.6	1.1	0.0–7.8	30.2
60 +	10,861	1.9	1.2	0.0–9.1	34.9
Gender
Female	14,812	2.5	1.5	0.0–14.0	47.6
Male	16,332	2.8	1.6	0.1–12.2	52.4

Fig. 1.  Seasonal pattern in daily mortality, daily mean temperature and precipitation in the Kassena-Nankana Districts (KNDs) of Northern Ghana (1995–2010).

Table 2. Summary statistics of daily temperature and precipitation in the Kassena-Nankana Districts (KNDs) of Northern Ghana (1995–2010)

Variable	Mean	SD	Min	Max	25th	75th	% missing
Temperature (°C)
Maximum	35.3	3.8	20.4	44.2	32.5	35.6	4.0
Minimum	23.0	2.8	12.1	33.0	21.3	23.0	4.0
Mean	29.2	2.7	21.4	37.0	27.3	30.8	4.0
Precipitation (mm)	3.0	9.3	0.0	93.0	0.0	0.0	5.5

Fig. 2.  Relative risks (RRs) of daily mortality among age and sex groups with daily mean temperature over lagged strata. Gray regions are corresponding 95% confidence intervals.

Table 3. Percent increase (95% confidence interval [CI]) for all cause daily mortality associated with 1°C increase in mean daily temperature

Variable	Effect estimates
Temperature (C)	25th percentile		75th percentile
	% increase	95% CI	% increase	95% CI
All cause
Lag 0–1	0.19	(0.05, 0.21)	1.14	(0.12, 1.54)
Lag 2–6	0.75	(−0.53, 1.94)	0.32	(0.16, 0.25)
Lag 7–13	0.06	(−1.21, 1.95)	0.31	(0.14, 0.26)
Males
Lag 0–1	0.22	(0.03, 0.28)	1.11	(−0.23, 2.04)
Lag 2–6	0.82	(−0.86, 2.56)	0.43	(0.21, 0.34)
Lag 7–13	−0.04	(−1.72, 2.58)	0.43	(0.21, 0.34)
Female
Lag 0–1	0.15	(−0.03, 0.28)	1.16	(−0.23, 2.11)
Lag 2–6	0.67	(−1.06, 2.64)	0.20	(−0.03, 0.35)
Lag 7–13	0.18	(−1.55, 2.66)	0.17	(−0.06, 0.35)
Age groups 0–4
Lag 0–1	0.05	(−0.19, 0.36)	0.61	(−1.20, 2.78)
Lag 2–6	−0.13	(−2.37, 3.48)	0.67	(0.36, 0.47)
Lag 7–13	1.14	(−1.14, 3.51)	0.50	(0.19, 0.47)
Age groups 5–19
Lag 0–1	0.34	(−0.05, 0.58)	2.35	(−0.40, 4.20)
Lag 2–6	3.08	(−0.40, 5.31)	0.38	(−0.05, 0.66)
Lag 7–13	0.63	(−2.77, 5.33)	0.66	(0.22, 0.66)
Age groups 20–59
Lag 0–1	0.14	(−0.09, 0.35)	0.52	(−1.11, 2.50)
Lag 2–6	0.33	(−1.71, 3.15)	0.19	(−0.08, 0.40)
Lag 7–13	−0.39	(−2.42, 3.17)	0.14	(−0.13, 0.41)
Age groups 60 +
Lag 0–1	0.29	(0.07, 0.33)	1.52	(−0.09, 2.45)
Lag 2–6	1.09	(−0.92, 3.08)	0.12	(−0.14, 0.40)
Lag 7–13	−0.63	(−2.62, 3.10)	0.19	(−0.07, 0.40)

Note: Bold figures show statistical significant results.

Table 4. Cumulative lag effect as a percent increase (95% confidence interval [CI]) for all cause mortality associated with 1°C increase in mean daily temperature

	Cumulative effect at 25th percentile for all lags		Cumulative effect at 75th percentile for all lags
Category	% increase	95% CI	% increase	95% CI
All cause	1.00	(−0.82, 2.84)	1.78	(0.73, 2.85)
Male	0.99	(−1.39, 3.43)	1.99	(0.60, 3.40)
Female	1.00	(−1.45, 3.52)	1.54	(0.11, 2.99)
Age groups 0–4	1.06	(−2.15, 4.37)	1.79	(−0.09, 3.71)
Age groups 5–19	4.08	(−0.87, 9.27)	3.42	(0.57, 6.35)
Age roups 20–59	0.08	(−2.80, 3.05)	0.85	(−0.83, 2.56)
Age groups 60 +	0.75	(−2.09, 3.68)	1.84	(0.18, 3.53)

Note: Bold figures show statistical significant results.

Fig. 3.  Relative risks (RRs) of daily mortality among children under 5 years of age with daily mean temperature over lagged strata. Gray regions are corresponding 95% confidence intervals.

Fig. 4.  Relative risks (RRs) of daily mortality among adults 60 years and above with daily mean temperature over lagged strata. Gray regions are corresponding 95% confidence intervals.

Fig. 5.  Relative risks (RRs) of daily mortality among all age and sex groups with daily precipitation over lagged strata. Gray regions are corresponding 95% confidence intervals.

Table 5. Percent increase (95% confidence interval [CI]) for all cause daily mortality associated with 10 mm increased in precipitation

Variable	Linear effect
Precipitation (mm)	%	95% CI
All cause
Lag 0–1	1.19	(−0.46, 2.87)
Lag 2–6	1.71	(0.10, 3.34)
Lag 14–27	1.22	(−0.35, 2.81)
Males
Lag 0–1	1.79	(−0.39, 4.00)
Lag 2–6	2.92	(0.80, 5.09)
Lag 14–27	2.35	(0.28, 4.45)
Female
Lag 0–1	0.55	(−1.67, 2.82)
Lag 2–6	0.38	(−1.77, 2.59)
Lag 14–27	−0.02	(−2.13, 2.13)
Age groups 0–4
Lag 0–1	0.42	(−2.39, 3.32)
Lag 2–6	1.36	(−1.34, 4.13)
Lag 14–27	1.84	(−0.75, 4.51)
Age groups 5–19
Lag 0–1	−3.04	(−7.63, 1.78)
Lag 2–6	0.42	(−4.07, 5.12)
Lag 14–27	0.33	(−4.15, 5.03)
Age groups 20–59
Lag 0–1	1.86	(−0.90, 4.70)
Lag 2–6	2.09	(−0.62, 4.88)
Lag 14–27	−0.24	(−2.90, 2.50)
Age groups 60 +
Lag 0–1	2.11	(−0.43, 4.70)
Lag 2–6	1.92	(−0.57, 4.47)
Lag 14–27	1.78	(−0.64, 4.27)

Fig. 6.  Relative risks (RRs) of daily mortality among males with daily precipitation over lagged strata. Gray regions are corresponding 95% confidence intervals.