Effect of off-hours admission on in-hospital mortality in sub-Saharan Africa: a systematic review and meta-analysis Authors: Tilahun Degu Tsega, Almaw Genet Yeshiwas, Tewodros Worku Bogale, Chalachew Yenew, Gashaw Melkie Bayeh, Ahmed Fentaw Ahmed, Meron Asmamaw Alemayehu, Abathun Temesgen, Wei Ma Published: 2026 ABSTRACT BackgroundThe findings of previous works of literature on the effects of off-hours admission on in-hospital mortality in sub-Saharan Africa were inconsistent and inconclusive. However, there is paucity of information as single summarized finding. Therefore, this review aimed to determine the pooled effects of off-hours admission on the in-hospital mortality in sub-Saharan Africa.MethodsA systematic review and meta-analysis were conducted using studies from PubMed, MEDLINE, Cochrane Library, Web of Science, African Journals Online, and Google Scholar up to 30/12/2025. Study quality, heterogeneity and publication bias were assessed accordingly. Pooled estimates were estimated using a random-effects model. Subgroup and sensitivity analyses were performed. Odds ratios with 95% confidence intervals were used to quantify the association between off-hours admission and in-hospital mortality.ResultsThis review included 14 primary studies. The pooled magnitude of off-hours in-hospital mortality among all hospital admissions in sub-Saharan Africa was 12.39% (95% CI: 9.76, 15.02%). The nighttime mortality among the nighttime admissions was 26.1% (95%CI: 18.6, 33.6%). Besides, the pooled weekend’s in-hospital mortality among weekends’ admissions in sub-Saharan Africa was 31.5% (95% CI: 23.5, 39.5%). The pooled effects of off-hours admission on in-hospital mortality was higher for weekend admission (OR = 1.22, 95% CI: 1.01, 1.47) and nighttime admission (OR = 2.36, 95% CI: 1.40, 4.00) compared with weekday admission.ConclusionsOff-hours admission may be associated with increased mortality; however, the findings should be interpreted cautiously due to substantial heterogeneity and variability in study definitions.RegistrationThe protocol was registered at PROSPERO with reg. no “CRD42024621339”.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s41043-026-01377-9. Contributions to the literature This systematic review and meta-analysis provide new and updated scientific evidence for policymakers, healthcare professionals, and researchers concerning the association between out-of-working hours hospital admissions and in-patient mortality.Hospital admissions during nighttime, weekends, and holidays may be associated with increased in-patient mortality.Inadequate staffing of healthcare professionals and financial constraints in sub-Saharan Africa exacerbate in-hospital mortality during these periods as evidences revealed from the systematic review of the included published articles.The out-of-working hours in-hospital mortality rate generated in this review is derived from available primary studies conducted within sub-Saharan Africa and represents a general overview of the in-hospital mortality in sub-Saharan Africa’s healthcare context. This systematic review and meta-analysis provide new and updated scientific evidence for policymakers, healthcare professionals, and researchers concerning the association between out-of-working hours hospital admissions and in-patient mortality. This systematic review and meta-analysis provide new and updated scientific evidence for policymakers, healthcare professionals, and researchers concerning the association between out-of-working hours hospital admissions and in-patient mortality. Hospital admissions during nighttime, weekends, and holidays may be associated with increased in-patient mortality. Hospital admissions during nighttime, weekends, and holidays may be associated with increased in-patient mortality. Inadequate staffing of healthcare professionals and financial constraints in sub-Saharan Africa exacerbate in-hospital mortality during these periods as evidences revealed from the systematic review of the included published articles. Inadequate staffing of healthcare professionals and financial constraints in sub-Saharan Africa exacerbate in-hospital mortality during these periods as evidences revealed from the systematic review of the included published articles. The out-of-working hours in-hospital mortality rate generated in this review is derived from available primary studies conducted within sub-Saharan Africa and represents a general overview of the in-hospital mortality in sub-Saharan Africa’s healthcare context. The out-of-working hours in-hospital mortality rate generated in this review is derived from available primary studies conducted within sub-Saharan Africa and represents a general overview of the in-hospital mortality in sub-Saharan Africa’s healthcare context. Introduction Off-hours admission is the hospital admissions time that occur outside of regular working hours, which may include nights, lunch hours, weekends, public holidays and or calendar days [1, 2]. The off-hours effect refers to differences in patient outcomes associated with care delivered outside regular weekday working hours [3]. Off-hours typically include nights, weekends, and public holidays, when staffing, senior supervision, and service availability may be reduced. It describes the higher risk of death among patients admitted outside regular working hours [2, 4]. This has been widely studied and documented since the 1970s particularly in developed countries. Studies have demonstrated poorer outcomes for patients hospitalized or treated during off-hours [5, 6]. Recent studies and meta-analyses have also reported this effect [7–10]. Although off-hours admissions have been associated with higher mortality, evidence suggests that improved care protocols can mitigate this risk. This indicates that systemic factors may be more influential than admission timing alone [11–15]. Some studies on specific patient groups, like those with heart failure and stroke, found no difference in mortality between off-hours and weekday admissions [16, 17]. The reasons for the weekend effect remain debated. Potential explanations include patient illness severity and the presence of comorbidities [18, 19]. Other contributing factors may include staffing levels during off-hours, delays in necessary interventions, lack of supervision, and inconsistent quality of care [20, 21]. Research in sub-Saharan Africa showed that trauma patients admitted at night had a higher mortality rate (1.4%) than those admitted during the day (0.8%). This suggests a potential increase in mortality associated with off-hours admissions [22]. In Uganda, patients presenting at night or on weekends had a higher likelihood of dying within 24 h of arrival [12]. Patients admitted during off-hours in Ghana experienced lower mortality rates, with 17% mortality during these time compared to 27% during working-hours [23]. On the other hand, a study on acute ST-Elevation Myocardial Infarction patients found no notable difference in mortality rates between admissions during working hours and off-hours [24]. A study in Ethiopia found that weekend admissions resulted in a 38% higher mortality rate in medical wards and 50% in intensive care units (ICUs) compared to weekdays [25]. In Sub-Saharan Africa region, the effect of off-hours admission on in-hospital mortality shows a complex relationship. This relationship is influenced by factors such as admission timing and the nature of medical care provided [26–28]. A systematic review and meta-analysis found that patients with nonvariceal upper gastrointestinal hemorrhage admitted during off-hours had higher mortality rates. These rates were significantly higher compared with patients admitted during regular hours [29]. However, this study found that off-hours admission was not associated with an increased risk of rebleeding. This suggests that the timing of admission may not influence this specific outcome [29]. Many studies have shown that patients admitted during off-hours, particularly on weekends and at night, have higher mortality rates [30–32]. On the other hand, works of literature in sub-Saharan Africa had an inconclusive finding on the effects of off-hours admission on the in-hospital mortality [27, 28, 33–35]. Some studies indicate that off-hours admission is significantly associated with in-hospital mortality. However, other studies suggest that off-hours admission may not affect in-hospital outcomes in Sub-Saharan Africa. Given these discrepancies, a clear synthesis of evidence specific to Sub-Saharan Africa is needed. This will help better understand the overall effect of off-hours admissions on in-hospital mortality. Therefore, this study aimed to estimate the pooled effect of off-hours admission (nighttime and weekends) on in-hospital mortality in sub-Saharan Africa. Methods Study settings A systematic review and meta-analysis of published articles in sub-Saharan Africa were conducted. This review attempted to search unpublished studies or grey literature from University library repositories, institutional catalogues and research square sources. However, none were accessible. The pooled magnitude of in-hospital mortality of patients on weekends’, holidays, or nighttime admissions was estimated. The effects of weekends’, holidays or nighttime admissions on the in-hospital mortality in sub-Saharan Africa was also determined. Search strategy Adequate literature search was conducted to assess potential duplication of existing reviews using multiple databases and the PROSPERO registry. Nothing was found and the protocol was registered at PROSPERO with registration number “CRD42024621339”. Then, this review was done using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 (PRISMA) guidelines [36] (Supplementary Material 1: S1). ((“off-hours” OR “out-of-hours” OR “after-hours” OR “nighttime” OR “weekend*” OR “weekend effect” OR “off-hour admission” OR “out of hours care”) AND (“in-hospital mortality” OR “hospital mortality” OR mortality OR death OR “clinical outcome*”) AND (“sub-Saharan Africa”)) (Supplementary Material 1: S2) was the general search strategy. Using these search terms, all sub-Saharan African countries were included. And ‘MeSH’ terms were also applied in databases to identify published articles and grey literature. PubMed, Cochrane library, MEDLINE, African Journals Online (AJOL), Web of Science and google scholar were the databases assessed to access the articles. Searching was done either in combination or independently using the search terms in the above databases accordingly. However, access to some databases was limited due to subscription restrictions. Study selection: inclusion and exclusion criteria All studies either published or grey literature done before 30/12/2025 were included in this review. All observational studies, national surveys, and audits reported at least the effect of weekends’, holidays or nighttime admissions on the in-hospital mortality were included. All these studies involved hospitalized patients of any age. This broad inclusion criteria (patients of any age and all causes of admission) used to ensure comprehensive capture of evidence on the off-hours effect across diverse clinical populations. While this approach increases clinical and methodological heterogeneity, it was considered appropriate for evaluating the overall association between off-hours admission and in-hospital mortality. Hence, the aim of this review was to identify the effect of weekends’, holidays or nighttime admissions on the in-hospital mortality, studies that reported only hospital mortality without these factors were also excluded. Although we planned to exclude non-English articles, none were identified during the search. Operational definition In-hospital mortality: is the death or mortality of patients after admitted or arrived into the hospital either in weekday, nighttime, weekend day or holiday. The systematic review section of this review compared the hospital mortality rate of weekday; and weekend day, holiday or nighttime admission. The pooled magnitude of in-hospital mortality among patients admitted during weekends, holidays, or nighttime was estimated using meta-analysis, irrespective of the cause of admission. Weekend hours admission: is the admission or arrival of patients in any cause of morbidity during Saturday or Sunday’s or holiday such as Easter, Christmas or other public and calendar days (may be from Monday to Friday). This review included primary studies that used this operational definition. Weekday hours admission: is the admission or arrival of the patients to the hospital on Monday to Friday (when these days were not holiday) from 8:00 AM to 5:30 PM. This review included studies that measured weekdays’ admission in this way. Nighttime hours admission: is the admission or arrival of the patients to the hospital on Monday to Sunday from 5:30 PM to 7:59AM. This review included primary studies that measured night times’ admission in this way. Off-hours’ admission: is the admission of patients to the hospital on either of weekend, nighttime, holiday, or calendar day. This review included studies that measure this variable in either of this way. Off-hours’ in-hospital mortality: is the death of patients who were admitted to, or presented at a hospital during weekends, nighttime, or holidays. It is calculated as the number of deaths among patients admitted during off-hours divided by the total number of admissions during those periods. However, for this review, the overall pooled estimate of off-hours in-hospital mortality was calculated by considering the total number of hospital admissions across all included studies as the denominator. Primary outcome variable: the effect of off-hours admission on in-hospital mortality in sub-Saharan Africa as compared with weekday admissions. Secondary outcome variable: the rate of in-hospital mortality among off-hours admission in sub-Saharan Africa. Data extraction and quality assessment After screening the studies based on their titles, abstracts, and full-text reviews, the selected articles were included in the systematic review and meta-analysis. All member of the research team together developed a predefined data extraction tool (Supplementary Material 1: S3) to ensure consistency. The data extraction tool contains author-date, study area details, publication status, study design and population, in-hospital mortality on each exposure variable, and the sample size (Supplementary Material 1: S3). The authors were organized into four teams (TDT &AGY, CHY&GMB, AFA&TWB, WA, MAA &AT) to extract data using a predefined tool. All discrepancies were resolved through discussion among all members of the research team. The quality of each studies were assessed using Newcastle-Ottawa Scale (NOS) quality assessment checklist for cohort and case control studies (Supplementary Material 1: S4) by one of the team [37], and the Joanna Briggs Institute critical appraisal tool (JBI) [38] for the cross-sectional studies (Supplementary Material 1: S5) by another of the team. The studies were assessed by each domain of the criterion as explicitly attached in the Supplementary Material 1: (S4) and (S5). Studies were classified as good quality for descriptive purposes if their quality score was ≥ 7, based on the mean score across included studies. This cutoff was used only to generate a categorical variable for descriptive and subgroup analyses and was not a validated standard. Sensitivity analyses excluding studies with scores < 7 showed no significant change in the pooled estimates. Heterogeneity The magnitude of the heterogeneity of the included studies was evaluated using the I² statistics whereas the statistical significance was assessed using Q-statistics. A p-value of less than 0.05 from the Q-statistic was considered indicative of statistical heterogeneity. Heterogeneity levels were categorized as low, moderate, or high (substantial), corresponding to I² values of 25%, 50%, and 75%, respectively [39]. Galbraith plot was also used to identify outliers that would contribute for heterogeneity. Publication bias Publication bias was examined using the Egger regression test [40] and Begg’s correlation test [41]. If publication bias was detected (p < 0.05), the trim-and-fill analysis method with a random-effects model was applied to adjust for the bias [42]. However, there was no evidence of statistically significant publication bias detected in this review. Data management and analysis methods The data were initially entered into Microsoft Excel 2019, and meta-analysis was performed using STATA 17 software (S6). Forest plots were utilized to show the pooled magnitude of in-hospital mortality on weekends’, holidays or nighttime admissions. Forest plots is also used to present the effect of weekend, holidays, or nighttime admissions on the in-hospital mortality in Sub-Saharan Africa. To address the heterogeneity among included studies, a DerSimonian Laird (DL) random-effects model was employed for the analysis [39]. Subgroup analyses were also conducted based on various study characteristics, including study design (cross-sectional or retrospective), study population and publication year. Additionally, meta-regression analysis was performed to explore potential sources of heterogeneity by considering factors such as publication year, study design, study quality score, region, study area, study population and country, where the study was conducted. However, only two (study population and publication year) significant variables that could account for the heterogeneity was discovered by meta-regression analysis. The effects of weekend, holidays, or nighttime admissions on the in-hospital mortality were expressed as odds ratios (ORs) with 95% confidence intervals (CIs). To enable pooled analysis, effect estimates reported in terms of relative risk (RR) or hazard ratios (HR) were converted to OR using standard approximation methods under the rare outcome assumption (OR ≈ RR). We acknowledge that this conversion may introduce bias, particularly when event rates are not rare. Sensitivity analyses restricted to studies reporting ORs, as well as analyses including converted RR/HR estimates, produced consistent results. Results Study selection A total of 146 records were initially identified through electronic database searches. After removing 98 duplicate records, 48 articles remained for screening based on their titles, abstracts, and full-text reviews. Of these, 34 articles were excluded during the screening process. Ultimately, 14 full-text articles met the inclusion criteria (Supplementary Material 1:S7) (Fig. 1). Fig. 1PRISMA flow diagram of included studies in the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in Sub-Saharan Africa until Dec.30/2025 PRISMA flow diagram of included studies in the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in Sub-Saharan Africa until Dec.30/2025 Systematic review and characteristics of included studies In this review, 14 primary published studies from seven sub-Saharan African countries at different region of each country were included [14, 15, 25–28, 33, 34, 43–48]. Most studies were conducted in teaching and tertiary public hospitals, although a few were conducted at public and private referral hospitals. Around 42% (six in number) and 20% (three in number) of the included studies were conducted in Ethiopia [25, 26, 28, 44, 46, 48] and Nigeria [27, 33, 43] respectively. Out of all primary studies included in this review, eight (three of them were prospective) of them were cohort studies [14, 15, 25, 26, 28, 34, 43, 48]. In addition, five cross-sectional [27, 33, 36, 44, 45] and one case control [46] studies were also included (Table 1). Among the primary studies that reported the effect of nighttime admission, approximately 43% found no statistically significant association between nighttime admission and in-hospital mortality [27, 33, 46]. Similarly, among the primary studies that reported the effect of weekend admission, approximately 50% of them found no statistically significant difference in the in-hospital mortality between weekend and weekday admissions [14, 15, 27, 45]. Among the included studies, seven reported either a positive or a negative association between nighttime admission and in-hospital mortality [26, 27, 33, 43, 44, 46, 47]. Whereas eight studies reported an association between weekend admission and in-hospital mortality [14, 15, 25, 28, 33, 44, 45, 48]. However, all 14 included studies reported the in-hospital mortality of patients among the total hospital admissions (Table 1). All included hospital-based primary studies reported a total sample size of 20,425 participants. The sample size ranges from a minimum 267 in a study conducted in Osun state, Nigeria [27] to a maximum of 5564 in a study conducted in Addis Ababa, Ethiopia [25]. Descriptive analysis of the included studies showed that 4,890 (23.94%) patients were admitted on weekends and 6,596 (32.29%) were admitted during nighttime hours. Of the 2,755 total in-hospital deaths (13.49%) reported across the primary studies, 27.19% (n = 749) occurred among weekend admissions and 19.96% (n = 550) among nighttime admissions. Together, these accounted for 6.36% (95% CI: 6.03, 6.69) of deaths among total hospital admissions. Specifically, 15.32% of patients admitted on weekends and 8.34% of those admitted during nighttime hours died. In contrast, there were 13,174 (64.49%) weekday admissions, among which 1,503 deaths were recorded, representing 54.56% of all in-hospital deaths. Based on the included studies, the prevalence of weekday mortality among total admissions in sub-Saharan Africa was 7.36% (95% CI: 7.00, 7.73). Overall, descriptive analysis indicated that the proportion of deaths was slightly higher among weekday admissions compared with off-hours (weekend and nighttime) admissions. Table 1Summary of studies’ characteristics on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Publication year)Study areaStudy designSample sizeTotal mortalityStudy populationWeekends’ admissionWeekends’ mortalityNighttime admissionNighttime mortalityPrevalence of off-hours mortality1Quality scoreAbate, et al. (2023) [26]Southern EthiopiaProspective cohort630303ICUMV* patients’101651908924.44%8Ansa V, et al. (2016) [43]South Easter NigeriaRetrospective cohort339107Cardiovascular patients82171413515.34%7Arega B, et al. (2024) [25]Addis Ababa, EthiopiaRetrospective cohort5564817Adult medical and ICU patients17423052601-5.48%8Demisse, et al. (2021) [44]Oromia, EthiopiaCross-sectional38172Road traffic accident8821762311.55%8Gathara, et al. (2013) [45]Nairobi, KenyaCross-sectional2901342Pediatric pneumonia patients996126--4.34%6Mediratta, et al. (2016) [46]Amhara, EthiopiaCase-control812207NICU** patients--47012915.89%7Mgaya, et al. (2021) [47]Dares Salaam, TanzaniaCross-sectional2636112Neonates among live births--1279692.62%8Shumet and Kebede. (2022)[48]Amhara, EthiopiaRetrospective cohort384150Adult ICU patients139641526633.85%7G.S.Stone, et al. (2012) [15]Western KenyaRetrospective cohort956211Medical patients26155--5.75%6Teshager NW, et al. (2019) [28]Amhara, EthiopiaProspective cohort313102Pediatrics ICU patients7231--9.90%7Eze P, et al. (2018) [33]South east, NigeriaCross-sectional455634Neonates among live births1266-129180.176%8Ndegbu CU, et al. (2015) [27]Osun state NigeriaCross-sectional26735Abdominal surgery patients--106134.87%8Prin, et al. (2018) [14]Central Lilongwe, MalawiProspective cohort377214Adult ICU patients104581679640.85%7Twahirwa I, et al. (2022) [34]RwandaRetrospective cohort30949Emergency laparotomy patients397123229.39%6ICUMV*= Intensive care unit mechanical ventilation, NICU**=Neonatal intensive care Unit, 1 = calculated from the total sample size (All admissions) Table 1Summary of studies’ characteristics on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Publication year)Study areaStudy designSample sizeTotal mortalityStudy populationWeekends’ admissionWeekends’ mortalityNighttime admissionNighttime mortalityPrevalence of off-hours mortality1Quality scoreAbate, et al. (2023) [26]Southern EthiopiaProspective cohort630303ICUMV* patients’101651908924.44%8Ansa V, et al. (2016) [43]South Easter NigeriaRetrospective cohort339107Cardiovascular patients82171413515.34%7Arega B, et al. (2024) [25]Addis Ababa, EthiopiaRetrospective cohort5564817Adult medical and ICU patients17423052601-5.48%8Demisse, et al. (2021) [44]Oromia, EthiopiaCross-sectional38172Road traffic accident8821762311.55%8Gathara, et al. (2013) [45]Nairobi, KenyaCross-sectional2901342Pediatric pneumonia patients996126--4.34%6Mediratta, et al. (2016) [46]Amhara, EthiopiaCase-control812207NICU** patients--47012915.89%7Mgaya, et al. (2021) [47]Dares Salaam, TanzaniaCross-sectional2636112Neonates among live births--1279692.62%8Shumet and Kebede. (2022)[48]Amhara, EthiopiaRetrospective cohort384150Adult ICU patients139641526633.85%7G.S.Stone, et al. (2012) [15]Western KenyaRetrospective cohort956211Medical patients26155--5.75%6Teshager NW, et al. (2019) [28]Amhara, EthiopiaProspective cohort313102Pediatrics ICU patients7231--9.90%7Eze P, et al. (2018) [33]South east, NigeriaCross-sectional455634Neonates among live births1266-129180.176%8Ndegbu CU, et al. (2015) [27]Osun state NigeriaCross-sectional26735Abdominal surgery patients--106134.87%8Prin, et al. (2018) [14]Central Lilongwe, MalawiProspective cohort377214Adult ICU patients104581679640.85%7Twahirwa I, et al. (2022) [34]RwandaRetrospective cohort30949Emergency laparotomy patients397123229.39%6ICUMV*= Intensive care unit mechanical ventilation, NICU**=Neonatal intensive care Unit, 1 = calculated from the total sample size (All admissions) Summary of studies’ characteristics on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 ICUMV*= Intensive care unit mechanical ventilation, NICU**=Neonatal intensive care Unit, 1 = calculated from the total sample size (All admissions) Test of heterogeneity The heterogeneity of the studies included in this review was substantial, as indicated by the I² statistic (I² = 99.55%, Q statistic (Q [13] = 2913.53, P < 0.001)). This high level of heterogeneity may reduce the reliability of the pooled estimates, warranting cautious interpretation. However, Galbraith plot inspection did not identify any single study as a dominant outlier. This suggests that the heterogeneity was likely distributed across studies rather than driven by a single influential study (Fig. 2). Fig. 2Galbraith plot on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Galbraith plot on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Pooled magnitude of hospital mortality Overall in-hospital mortality rate Using random effects model, the pooled in-hospital mortality across all-causes admissions in sub-Saharan Africa was 23.5% (95%CI: 18.2, 28.7%). Based on the Z-test of theta result, overall pooled mortality rate was significantly different from zero (z = 8.76, p < 0.001) (Table 2). Table 2The overall in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Year)Rate of mortality (%) (95% CI)Weights (%)Abate, et al. (2023)49.0 (45.1, 52.9)7.10Ansa V, et al. (2016)31.6 (26.6, 36.5)6.93Arega B, et al. (2024)11.5 (10.7, 12.3)7.37Demisse, et al. (2021)18.9 (15.0, 22.8)7.09Gathara, et al. (2013)11.8 (10.6, 13.0)7.36Mediratta, et al. (2016)25.5 (22.5, 28.5)7.21Mgaya, et al. (2021)4.2 (3.4, 5.0)7.38Shumet and Kebede. (2022)39.1 (34.2, 43.9)6.94G. S. Stone, et al. (2012)22.1 (19.5, 24. 7)7.25Teshager NW, et al. (2019)32.6 (27.4, 37. 8)6.89Eze P, et al. (2018)0.7 (0.5, 1.01)7.39Ndegbu CU, et al. (2015)13.1 (9.1, 17.2)7.08Prin, et al. (2018)56.8 (51.8, 61.8)6.92Twahirwa I, et al. (2022)15.9 (11.8, 20.0)7.07Overall mortality (theta)23.50 (18.2, 28.7)Test of theta = 0: z = 8.76, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 2913.53, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 99.55%, H2 = 224.12 Table 2The overall in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Year)Rate of mortality (%) (95% CI)Weights (%)Abate, et al. (2023)49.0 (45.1, 52.9)7.10Ansa V, et al. (2016)31.6 (26.6, 36.5)6.93Arega B, et al. (2024)11.5 (10.7, 12.3)7.37Demisse, et al. (2021)18.9 (15.0, 22.8)7.09Gathara, et al. (2013)11.8 (10.6, 13.0)7.36Mediratta, et al. (2016)25.5 (22.5, 28.5)7.21Mgaya, et al. (2021)4.2 (3.4, 5.0)7.38Shumet and Kebede. (2022)39.1 (34.2, 43.9)6.94G. S. Stone, et al. (2012)22.1 (19.5, 24. 7)7.25Teshager NW, et al. (2019)32.6 (27.4, 37. 8)6.89Eze P, et al. (2018)0.7 (0.5, 1.01)7.39Ndegbu CU, et al. (2015)13.1 (9.1, 17.2)7.08Prin, et al. (2018)56.8 (51.8, 61.8)6.92Twahirwa I, et al. (2022)15.9 (11.8, 20.0)7.07Overall mortality (theta)23.50 (18.2, 28.7)Test of theta = 0: z = 8.76, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 2913.53, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 99.55%, H2 = 224.12 The overall in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Test of theta = 0: z = 8.76, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 2913.53, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 99.55%, H2 = 224.12 Overall weekends' in-hospital mortality rate Using random effects model (I² = 96.46%), the pooled weekends’ in-hospital mortality rate among weekends’ admissions in sub-Saharan Africa was 31.5% (95%CI: 23.5, 39.5%). Based on the Z-test of theta result, overall pooled mortality rate was significantly different from zero (z = 7.75, p < 0.001) (Fig. 3). Fig. 3The overall weekends’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 The overall weekends’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Overall nighttime in-hospital mortality rate Using random effects model (I² = 99.49%), the pooled nighttime in-hospital mortality rate among nighttime admissions in sub-Saharan Africa was 26.1% (95%CI: 18.6, 33.6%). Based on the Z-test of theta result, overall pooled mortality rate was significantly different from zero (z = 6.83, p < 0.001) (Table 3). Table 3The overall nighttime in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Year)Rate of mortality (%) (95% CI)Weights (%)Abate, et al. (2023)46.80 (39.70, 53.90)9.84Ansa V, et al. (2016)24.80 (17.70, 32.00)9.83Demisse, et al. (2021)30.30 (19.90, 40.60)8.96Mediratta, et al. (2016)27.40 (23.40, 31.50)10.47Mgaya, et al. (2021)5.40 (4.2, 6.6)10.76Shumet and Kebede. (2022)43.40 (35.5, 51.30)9.64Eze P, et al. (2018)0.60 (0.2, 0.10)10.79Ndegbu CU, et al. (2015)12.30 (6.00, 18.50)10.04Prin, et al. (2018)57.50 (50.00, 65.00)9.74Twahirwa I, et al. (2022)17.90 (11.10, 24.70)9.92Overall mortality (theta)26.10 (18.60, 33.60)Test of theta = 0: z = 6.83, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 791.23, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 98.86%, H2 = 87.91 Table 3The overall nighttime in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Authors’ name (Year)Rate of mortality (%) (95% CI)Weights (%)Abate, et al. (2023)46.80 (39.70, 53.90)9.84Ansa V, et al. (2016)24.80 (17.70, 32.00)9.83Demisse, et al. (2021)30.30 (19.90, 40.60)8.96Mediratta, et al. (2016)27.40 (23.40, 31.50)10.47Mgaya, et al. (2021)5.40 (4.2, 6.6)10.76Shumet and Kebede. (2022)43.40 (35.5, 51.30)9.64Eze P, et al. (2018)0.60 (0.2, 0.10)10.79Ndegbu CU, et al. (2015)12.30 (6.00, 18.50)10.04Prin, et al. (2018)57.50 (50.00, 65.00)9.74Twahirwa I, et al. (2022)17.90 (11.10, 24.70)9.92Overall mortality (theta)26.10 (18.60, 33.60)Test of theta = 0: z = 6.83, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 791.23, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 98.86%, H2 = 87.91 The overall nighttime in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Test of theta = 0: z = 6.83, Prob > |z| < 0.001Test of homogeneity: Q = chi2(13) = 791.23, Prob > Q< 0.001Heterogeneity: tau2 = 0.01, I2 = 98.86%, H2 = 87.91 Overall off-hours in-hospital mortality rate Based on the random effects model, the overall pooled off-hours in-hospital mortality rate among total admissions in sub-Saharan Africa was 12.39% (95%CI: 9.76, 15.02%) (I² = 99.09%, P < 0.001). The Z-test of theta for this pooled prevalence showed that it was significantly different from zero (z = 9.23, p < 0.001) (Fig. 4). Based on the Galbraith plot, no study fell outside the 95% confidence interval. The interval included the null value, indicating no evidence of influential outliers (Fig. 5). Fig. 4The overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 The overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Fig. 5The Galbraith plot for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 The Galbraith plot for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 To further explore sources of heterogeneity, meta-regression and subgroup analyses were conducted. In the meta-regression, two study-level variables, publication year and study population, were identified as explaining the observed between-study variation (Table 4). Statistically significant residual heterogeneity indicated that the model did not fully explain the observed variability, suggesting the presence of unmeasured factors. Therefore, subgroup analyses were conducted using selected study-level variables that may be contributed to the high heterogeneity. Table 4Meta-regression analysis of the study level variables for the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Study level variablesCoefficientsStandard errorZ-valueP > [z][95%CI]Country-1.3081.30338-1.000.315-3.863, 1.246Publication year11.1354.321052.58 0.010 2.665,19.604Study design0.3741.503700.250.804-2.574, 3.321Study population-1.3330.49950-2.67 0.008 -2.312, 0. 356Test of residual homogeneity: Q_res = chi2 (9) = 527.06 Prob > Q_res < 0.0001 Table 4Meta-regression analysis of the study level variables for the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025Study level variablesCoefficientsStandard errorZ-valueP > [z][95%CI]Country-1.3081.30338-1.000.315-3.863, 1.246Publication year11.1354.321052.58 0.010 2.665,19.604Study design0.3741.503700.250.804-2.574, 3.321Study population-1.3330.49950-2.67 0.008 -2.312, 0. 356Test of residual homogeneity: Q_res = chi2 (9) = 527.06 Prob > Q_res < 0.0001 Meta-regression analysis of the study level variables for the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Test of residual homogeneity: Q_res = chi2 (9) = 527.06 Prob > Q_res < 0.0001 Subgroup analysis by country The country level subgroup analysis shown that there was a statistically significant difference in the pooled off-hours in-hospital mortality rate (Fig. 6). In Ethiopia, the pooled proportion of off-hours in-hospital mortality among the total admissions was 16.70% (95% CI:8.96, 24.45%), which was more than the pooled magnitude of the sub-Saharan Africa included countries. Fig. 6The country level subgroup analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 The country level subgroup analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Subgroup analysis by study design As shown in Fig. 7, there was a statistically significant difference in pooled off-hours in-hospital mortality across study designs among the included studies for total hospital admissions (Fig. 7). The pooled proportion of off-hours in-hospital mortality among total admissions in prospective cohort studies was 24.97% (95% CI: 8.87, 41.08%), which was higher than the overall pooled estimate and those observed in other study designs (Fig. 7). The overall test for group differences by study design proved a statistically significant difference in the pooled magnitude of in-hospital mortality among patients admitted during off-hours (Fig. 7). Fig. 7The study design level subgroup analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 The study design level subgroup analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Subgroup analysis by study population In subgroup analysis by study population, the pooled proportion of off-hours in-hospital mortality varied significantly across groups (test for subgroup differences: Q (3) = 26.30, p < 0.001). Adult ICU patients had the highest pooled mortality of 32.92% (95% CI: 23.10, 42.74), followed by surgical patients at 8.51% (95% CI: 4.43, 12.58) and medical patients at 8.03% (95% CI: 4.88, 11.18). Neonatal patients had the lowest pooled mortality at 6.15% (95% CI: 3.28, 9.03). Substantial heterogeneity was observed within all subgroups (I² > 80%), indicating considerable variability among studies (Fig. 8). Fig. 8Subgroup analysis by study population for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Subgroup analysis by study population for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Sensitivity analysis All included studies had point estimates that lay within the 95% confidence intervals of the sensitivity analysis. Therefore, the influence of small-study effects on the pooled estimates of in-hospital mortality associated with off-hours admission was considered as minimal (Fig. 9). Fig. 9Sensitivity analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Sensitivity analysis for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Publication bias Funnel plot The funnel plot showed that there was asymmetry of the included studies distribution in the funnel, gave a subjective evidence of publication bias (Fig. 10). However, the egger’s (β = 4.86, P = 0.441) and Begg’s test (Kendall’s score = 25.00, p = 0.1889) revealed that there was no statistically significant publication bias. Fig. 10Funnel plot for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Funnel plot for the overall off-hours’ in-hospital mortality rate on the systematic review and meta-analysis of the effect of off-hours admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Effect of weekends' admission on in-hospital mortality Among the 14 studies included in this meta-analysis, only eight of them reported the effect of weekend admissions on the in-hospital mortality in sub-Saharan Africa, compared with weekday admissions. The pooled effect of weekend admissions on the in-hospital mortality was significantly higher, with an odds ratio of 1.22 (95% CI: 1.01, 1.47) compared with weekday admissions (Fig. 11). Fig. 11Forest plot for the effect of weekend’s admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Forest plot for the effect of weekend’s admission on the in-hospital mortality in sub-Saharan Africa, until Dec.30/2025 Effects of nighttime admission on in-hospital mortality Among the 14 studies included in this meta-analysis, only seven of them were reported the effects of nighttime admission on the in-hospital mortality in sub-Saharan Africa. The pooled effect of nighttime admission on the in-hospital mortality in sub-Saharan Africa was 2.36 (OR = 2.36, 95%CI: 1.40, 4.00) times higher than weekdays admission (Fig. 12). Fig. 12Forest plot for the effect of nighttime admission on the in-hospital mortality rate in sub-Saharan Africa, until Dec.30/2025 Forest plot for the effect of nighttime admission on the in-hospital mortality rate in sub-Saharan Africa, until Dec.30/2025 Discussion This systematic review and meta-analysis found that the pooled magnitude of in-hospital mortality in the off-hours among the total admissions was 12.39% (95%CI: 9.76, 15.02%). This reflects that patients admitted during off-hours in this review were observed to have higher mortality rates compared to weekday admissions. This is supported by the previous systematic review and meta-analysis on 28 specific diseases and 251 cohorts conducted globally showed that off-hour admissions were significantly associated with increased mortality risks than the weekdays admission [3]. In sub-Saharan African countries, where human resources and financial constraints are common, the health care professionals allotted to work in the off-hours (weekends’ and night times’) are low [25, 49, 50]. The patient-healthcare professional ratio to deliver the health services in most sub-Saharan countries is below the standard during the off-hours [25, 49, 51, 52]. This may be associated with delays or prolonged time to receipt of expected hospital services during these periods [51, 52]. One plausible explanation is that, during off-hours, healthcare services may be prioritized for emergency cases, and some non-emergency patients may not receive care and may be asked to return later. This may be performed even at the triaged stage. However, this interpretation reflects hypothesized system-level constraints, such as staffing limitations or service discontinuity, rather than evidence directly assessed in the included studies [19]. This may be associated with challenges in disease management, which could coincide with emergency presentations and hospital admissions during off-hours. However, a retrospective cohort study from U.S. academic hospitals found that the apparent excess mortality associated with off-hours admission was largely attenuated after adjustment for patient severity. The off-hours effect may be attributable to differences in the clinical risk profile of patients presenting during these periods rather than to reduced quality of care [53]. This may suggest that the relationship between admission timing and in-hospital mortality is highly context-dependent and sensitive to analytical adjustment. The implications are that pooled estimates from systematic reviews should be interpreted cautiously, as off-hours effects may reflect a combination of patient selection, health system organization, and methodological differences rather than a universal deficit in nighttime or weekend care. Therefore, while causality cannot be inferred, future policy-oriented research could assess whether more equitable resource allocation across off-hours and weekdays is associated with more consistent service provision in sub-Saharan African healthcare systems. Substantial statistical heterogeneity was observed across included studies, may indicating marked variability in effect estimates beyond chance. This heterogeneity represents an important analytical challenge and limits the generalizability of the pooled estimates. Meta-regression and subgroup analyses explained part of this variability. And identifying study design, country, study population, and publication year as significant sources of heterogeneity. These findings indicate that the effect of off-hours admission on in-hospital mortality is context-dependent rather than uniform. Residual heterogeneity persisted after these analyses, may suggesting the influence of unmeasured clinical and health system factors. Variations in study population, disease severity, healthcare infrastructure, staffing levels during off-hours, and access to timely interventions likely contribute to the observed differences across settings in Sub-Saharan Africa. Similar residual heterogeneity has been reported in prior meta-analyses of weekend or off-hours effects, where between-study variability often reflects real-world differences in care delivery rather than methodological limitations alone [10, 54]. Although Galbraith plot inspection did not identify any single influential outlier study, this does not contradict the presence of extreme heterogeneity. Instead, it suggests that heterogeneity was distributed across studies rather than driven by isolated extreme estimates. Sensitivity analyses further showed that pooled estimates were stable after excluding individual studies, supporting internal consistency despite distributed substantial between-study variability. While the direction of the association between off-hours admission and in-hospital mortality appears consistent, the magnitude of effect varies widely across contexts. Therefore, pooled estimates should be interpreted with caution, and greater emphasis could be placed on subgroup-specific and contextual findings when informing policy or clinical practice. The pooled effects of weekends’ admission on the in-hospital mortality was 1.22 times higher than the weekday’s admission in sub-Saharan Africa. This finding is in line with the findings of the previous systematic review and meta-analysis, which reported that off-hour admissions were significantly associated with increased mortality risks than the weekdays admission [3]. This finding is also supported by the findings of the studies done in high income countries, and stated that off-hours admission had a high risk of mortality rate (RR = 1.19 ) than on-hours admission [10, 19]. It might be explained as most patients seeking health care services during the weekends’ hour are emergency or have a severe illness. This may be associated with the high mortality rate of patients during the weekend time. Another study also supported ours findings, which stated that weekends’ time admission had a slight higher effect on the in-hospital mortality than weekdays admission [32]. In addition, a prior systematic review and meta-analysis reported findings consistent with ours, demonstrating that off-hour admission was associated with a higher risk of short-term mortality (OR = 1.09). However, another previous large scale study reported that weekends’ admission couldn’t have a statistically significant difference in-hospital mortality of patients [17]. This may be due to the specificity of the inclusion criteria of the sample population. Ours systematic review and metanalysis included any causes of patients’ admission in opposite to the previous large-scale study which included only stroke patients. The present study also found that study population is one cause of the heterogeneity of the included studies to this review. It is clear that the cause of the patients’ admission to the hospital may have a difference for the in-hospital mortality particularly during weekends’ hour. This may be associated with the quality of healthcare service variation during this time [55]. Another systematic review and meta-analysis had an opposite finding from the present review findings [35]. This stated that there was no statistically significant weekend effect for the in-hospital mortality of pneumonia patients with an OR = 1.03 [35]. This discrepancy may associate with the study designs of the included studies to the review. Ours study included all observational studies including cross-sectional, and case-control in opposite to the previous review that included only retrospective cohort studies, most of them done in high and middle income countries [35]. The pooled effect of nighttime’s’ admission on in-hospital mortality was 2.36 times higher than the weekdays’ admission in sub-Saharan Africa. The present review findings’ is supported by the systematic review and meta-analysis findings of the previous studies, reported that nighttime admission had a higher mortality rates, particularly for conditions like upper gastrointestinal hemorrhage and other emergency conditions as compared to the weekdays’ admission [29, 56]. Another systematic review and meta-analysis studies however, found that there were no significant association between nighttime admissions and increased in-hospital mortality particularly intensive care unit and myocardial infarction patients [57, 58]. Based on these reviews, staffing intensity have more effect on the in-hospital mortality of patients than the nighttime admission. Another Evidence suggests that the association between nighttime admission and higher in-hospital mortality is not consistent across settings. A large multicenter ICU registry study from Japan reported that off-hours ICU admissions were associated with lower in-hospital mortality after adjustment for illness severity and hospital-level factors [59]. This finding may reflect context-specific organizational features, including standardized ICU staffing models, reduced elective workload, and more focused attention on critically ill patients at night [59]. Importantly, patient severity at presentation did not differ substantially between daytime and nighttime admissions, challenging the assumption that off-hours patients are inherently sicker in all settings. With these findings, this review could have some limitations. So, the result should be interpreted in consideration of these limitations. It included all hospitalized patients such as emergency, and non-emergency patients of any age. This may differ the pooled effect of the off-hour’s admission in the in-hospital mortality. So, the coming review studies’ better to consider this as the area of concern. Besides, unexplained sources of heterogeneity of the included studies was detected in this review, even though publication year and study population were among the sources of heterogeneity identified by meta-regression. Only studies published in English were accessed, which may have introduced language bias. The pooled magnitude of the in-hospital mortality in the off-hours was estimated among the total admissions. This may underestimate the magnitude. The effect of off-hours admission was also identified from the total hospitalized patients, this may underestimate the pooled effect. These may be the future review concern. This review gave more emphasis on those primary studies reported the effects of the weekends’ and nighttime admissions rather than the total in-hospital mortality, which may underestimate the pooled magnitude of in-hospital mortality of patients. Conclusion and recommendation In sub-Saharan Africa, in-hospital mortality for off-hours admissions stands at 12.39%, with weekend admissions carrying a 1.22 times higher risk and night-time admissions a 2.36 times higher risk compared to weekdays. However, the findings should be interpreted cautiously due to substantial heterogeneity and variability in study definitions. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1