The impact of COVID-19 and public health measures on homicide and suicide trends in Botswana using a
The impact of COVID-19 and public health measures on homicide and suicide trends in Botswana using an interrupted time series analysis

The impact of COVID-19 and public health measures on homicide and suicide trends in Botswana using an interrupted time series analysis

How did your country report this? Share your view in the comments.

Diverging Reports Breakdown

The impact of COVID-19 and public health measures on homicide and suicide trends in Botswana using an interrupted time series analysis – Injury Epidemiology

During the study period, A total of 6,094 medico-legal deaths were recorded in Botswana between January 2018 and September 2022. Males accounted for the majority of both homicides (62.7%) and suicides (88.2%). The mean age of homicide and suicide decedents was 32.4 (sd: 15.0) and 34.0 (SD: 12.3) years, respectively. For homicides, monthly autopsy counts declined during the SoE period compared with the pre-SoE period. For suicides, the median and spread of monthly autopsies did not appear to change very much. However, the monthly number of suicides increased during the post- soe period. Female suicides were very low compared with male suicides. Autopsies for male suicide victims steadily increased over theStudy period. There was some evidence to suggest that homicides of males were seen among the largest declines among males, but there was also evidence that suicides of females increased dramatically after the COVID emergency.

Read full article ▼
During the study period, A total of 6,094 medico-legal deaths were recorded in Botswana between January 2018 and September 2022. Information on these deaths was obtained from forensic pathology registers. Among these, 2,052 were classified as road traffic accidents, 1,479 as homicides, 746 as suicides, 832 as natural deaths, 564 as accidental deaths, and 421 as undetermined causes and represent 100% of all medico-legal deaths of this nature in the country. A total of 2,225 homicide and suicide autopsies were reviewed, comprising 1,479 homicides and 746 suicides. Males accounted for the majority of both homicides (62.7%) and suicides (88.2%). The mean age of homicide and suicide decedents was 32.4 (sd: 15.0) and 34.0 (sd: 12.3) years, respectively. See Table 1 for full results.

Table 1 Descriptive results of the study sample Full size table

Comparing the periods before, during, and after the state of emergency (SoE), the median number of autopsies was lowest during the SoE period. There were 40 (IQR: 15.8) autopsies per month before the SoE, 32 (IQR: 11.5) autopsies per month during the SoE, and 46 (IQR: 11.5) autopsies per month after the SoE. Similarly, when considering the rate of autopsies per 1,000,000 population, the median number was lower during the SoE period than during the pre- and post-SoE periods. For homicides, there were 28 (IQR: 7.5), 19 (IQR: 7), and 29.5 (IQR: 10) for each of the pre-SoE, SoE, and post-SoE periods, respectively. For suicides, there were 12.5 (IQR: 4), 13 (IQR: 6), and 16.5 (IQR: 4.75) for each of the pre-SoE, SoE, and post-SoE, respectively.

Boxplots illustrating the median and spread of monthly counts of homicides and suicides are shown in Figs. 1 and 2, respectively. For homicides, monthly autopsy counts declined during the SoE period compared with the pre-SoE period. Homicide autopsies returned to pre-SoE levels during the post-SoE period. This pattern was evident for both male and female sex groups and for the 0–17 year old and 18–45 year old age groups. For suicides, monthly autopsy counts declined during the SoE period compared with the pre-SoE period. Homicide autopsies returned to pre-SoE levels during the post-SoE period. This pattern was evident for both male and female sex groups and for the 0–17 year old and 18–45 year old age groups.

For suicides, the median and spread of monthly autopsies did not appear to change very much for the SoE period compared with the pre-SoE period. However, the monthly number of suicides increased during the post-SoE period compared with the pre-SoE and SoE periods. Female suicides were very low compared with male suicides. Autopsies for male suicide victims steadily increased over the study period. Most suicide autopsies were for people between the ages of 18 and 45 years. Following the SoE period, monthly counts of suicide autopsies for people over 45 years increased (See Fig. 2).

Fig. 1 Boxplots showing the median and spread of monthly counts of homicides in the pre-emergency, emergency, and post-emergency periods. Boxplots of monthly counts of homicides for each of the COVID emergency period are also presented for sex and age groups Full size image

Fig. 2 Boxplots showing the median and spread of monthly counts of suicides in the pre-emergency, emergency, and post-emergency periods. Boxplots of monthly counts of suicides for each of the COVID emergency period are also presented for sex and age Full size image

Fig. 3 Time series plots of monthly homicide and suicide autopsies for sex groups for the pre-emergency, emergency, and post-emergency periods Full size image

Figure 3 shows time series plots of monthly homicide and suicide autopsies, with the time series compared by sex and delineating the pre-SoE, SoE, and post-SoE periods. Monthly counts of homicide autopsies declined during the emergency period, but the largest declines were seen among males. There was some evidence to suggest that homicides of females declined soon after the onset of the COVID-19 emergency, but monthly counts soon began to increase. Following the COVID-19 emergency, female homicides returned to pre-pandemic levels, but male homicides increased dramatically. For suicides, there was little evidence to suggest changes in the temporal patterns of monthly autopsies. Visual evidence even suggests a slight increase among males.

Table 2 All estimates for interrupted time series analyses of homicide and suicide autopsies including estimates for the intercept, the slope (change in the number of monthly autopsies over time) of the pre-emergency period, the mean change of the number of autopsies in the emergency period compared with the pre-emergency period, the slop of the emergency period, the mean change in the number of autopsies in the post-emergency period compared with pre-emergency period and the slop of the number of autopsies over time for the post-emergency period Full size table

Interrupted time series analysis is of homicide and suicide autopsies during the presoe, soe, and post-SoE periods

We constructed models of monthly counts of homicide and suicide autopsies, including separate intercept and slope terms for each of the periods in a linear model. The results of these analyses are presented in Table 2 and in Figs. 4 (homicide) and 5 (suicide). Based on the time slope, there was no evidence to suggest that homicides overall were increasing during the pre-SoE period, but homicide autopsies for males increased with an extra 0.13 deaths per year per 100,000 men (95% CI: 0.06, 0.21). There was a clear statistically significant increase in monthly suicide autopsies before the SoE period with an excess 0.2 deaths per year per 100,000 population (95% CI:0.01, 0.38). This increase was most pronounced in females with an excess death of 0.12 deaths per year per 100,000 females (95% CI:0.08, 0.17). There was a slight decrease in suicides among very young people with a reduction of 0.05 deaths per year per 100,000 population (95% CI: −0.07, −0.04).

During the SoE period, there was a clear decrease in the overall number of autopsies for both homicide and suicide The overall homicides presented a statistically significant sudden drop after the implementation of the state of emergency, and in both sex and age groups, except those above 45 years. The biggest difference was in men and the age-group between 18 and 45 years, with a reduction of 9.92 deaths per 100,000 men per month (95% CI: −12.07, 7.77) and 11.33 deaths per 100,000 population per month (95% CI: −14.63, −8.02) respectively. As for suicides, there was also a statistically significant decrease immediately after the implementation of the state of emergency in overall, females, and the age-group between 18 and 45, but a slight increase in those above 45 years. An overall reduction of 4.54 (95% CI: −8.86,−0.21), 1.39 (95% CI: −2.47, −0.31), 3.48(95% CI: −5.92, −1.03) and an excess death of 1.05 (95% CI: 0.09, 2.01) respectively per 100,000 population per month.

Based on the time slope, during state of emergency, there was a statistically significant increase in female victims of homicide and adults aged 18–45 years. An excess of 0.37 excess death of 0.37 deaths per year per 100,000 females (95% CI: 0.21, 0.53) and an extra 0.23 deaths per year per 100,000 population (0.02,0.44) but no change for other groups. There was no evidence to suggest that there were increases or decreases in suicides overall or for any sex or age groups except very young people ((an excess of 0.06 deaths per year per 100,000 population) (95% CI: 0.04, 0.08)) during the SoE period.

A rise in under-17-year suicides was noted immediately after the lifting of the restrictions, an excess of 0.51 deaths per month per 100,000 population per month compared to pre-state of emergency. During the post-SoE period and the lifting of restrictions, there was an average return to pre-SoE monthly counts of homicides and suicides. There was no evidence to suggest any significant decrease or increase in average monthly counts of autopsies in the post-SoE period compared with the pre-SoE period. However, The time slope in the post-SoE was negative for all homicides except for the under 17 years age-group and the slope of all suicides was positive. We found that the temporal pattern of monthly autopsies for homicide following the lifting of restrictions declined in this period. A statistically significant difference was noted in the overall homicides, males, females and the age group 18–45 years: −1.24 (95% CI: −1.98, −0.51), −0.6 (95% CI: −1.01, −0.19), −0.6.2 (95% CI: −1.04, 10.19) and − 1.1(95% CI: −1.68, −0.51) respectively. For the temporal pattern of suicides following the lifting of restrictions, there was no evidence to suggest any change over time except within adults over the age of 45 years which showed an excess of 0.28 deaths per year per 100,000 population (95% CI: 0.11, 0.46).

Fig. 4 Interrupted time series analyses of monthly counts of autopsies of homicides. Black lines represent the intercepts and slopes of the linear model for the pre-emergency, emergency, and post-emergency periods. Dotted lines represent the predicted trajectory of monthly counts after each of the pre-emergency and emergency period assuming no change (no interruption). Colors represent the confidence intervals for slopes of change over time for each of the periods Full size image

Source: Injepijournal.biomedcentral.com | View original article

Source: https://injepijournal.biomedcentral.com/articles/10.1186/s40621-025-00592-y

Leave a Reply

Your email address will not be published. Required fields are marked *