Sana El Mhamdi

Adverse events in a Tunisian hospital: results of a retrospective cohort study

BACKGROUND Despite the worldwide growing attention to patient safety, Tunisia has no data on the magnitude and consequences of hospital adverse events (AEs). OBJECTIVE To estimate the incidence, nature and consequences of AEs and preventable AEs in a university hospital in Tunisia. DESIGN AND SETTING We opted for a two-stage retrospective medical record review of 620 inpatients admitted during 2005 based on the use of 18 screening criteria. Records were reviewed by a trained medical student, then by an expert physician when one or more criteria were identified. Main outcomes measures We determine the incidence, preventability and consequences of the AEs. Patients and admissions characteristics were also recorded. RESULTS Among 620 inpatients, 62 inpatients experienced an AE with an incidence of 10% (95% CI [7.6-12.3]). Surgical/invasive procedures and therapeutic errors were the most common AEs (55 and 21%, respectively). Among the confirmed events 60% were judged to be highly preventable and 21% led to patient death. All ages and both genders experienced equal rates of AEs. However, patients who experienced these events were significantly more exposed to extrinsic risk factors (all surgical interventions and invasive procedures that were listed in the revue form 2 of the questionnaire). Physician reviewers estimated that a total of 570 additional hospital days were associated with AEs. CONCLUSION This study confirms that preventable AEs were not rare in our context. They caused human harm and consumed a significant part of hospital resources. Thus, targeted interventions are needed.

Obstetric and psychological characteristics of women seeking multiple abortions in the region of Monastir (Tunisia): results of a cross-sectional design

BackgroundRepeat abortion is a public health concern favored by many obstetric and social factors.The purpose of our study was to identify associated factors to repeated abortion in the region of Monastir (Tunisia). Common mental disorders (CMD) such as anxiety and depression were also evaluated in women seeking voluntary repeated abortion.MethodsWe carried out a cross sectional study between January and April 2013 in the Reproductive Health Center (RHC) of the region of Monastir in Tunisia (This study is part of a prospective design on mental disorders and intimate partner violence among women seeking abortions in the RHC).Among women referred to the RHC we selected those seeking voluntary abortion (medical or surgical method). Data on women’s demographic characters, knowledge and practices about contraceptive methods and abortion were collected the abortion day via a structured questionnaire. Data about anxiety and depression status were evaluated during the post-abortal control visit at 3–4 weeks following pregnancy termination.ResultsOf the 500 interviewed women, 211 (42.2 %; CI95% [37.88 – 46.52]) were seeking repeated abortions. Multivariate analysis showed that increased age, lower level of women school education, single status, poor knowledge about birth control methods and history of conflict/abuse by a male partner, were uniquely associated with undergoing repeat compared with initial abortion.CMD were significantly higher in women undergoing second or subsequent abortion (51.1 %) single and lower educated women. Women relating a history of conflicts/abuse report more CMD than others (30.6 % vs 20.8 %).ConclusionHealth facilities providing abortion services need to pay more attention to women seeking repeat abortion. Further studies are needed to well establish the relation between the number of abortion and the occurrence and the severity of CMD.

Women’s knowledge, attitudes and practice about breast cancer screening in the region of Monastir (Tunisia)

Breast cancer remains a worldwide public health problem. In Tunisia, it is considered to be the primary womens cancer and causes high morbidity and mortality. This study aimed to investigate female knowledge, attitudes and practice of breast cancer screening in the region of Monastir (Tunisia). We conducted a descriptive cross-sectional design exploring knowledge, attitudes and practices of women in the region of Monastir on breast cancer screening. The study was conducted in health centres of this region from 1 March 2009 to 30 June 2009. Data were collected via a structured questionnaire containing 15 items on demographic status, knowledge of risk factors and screening methods and attitudes towards the relevance and effectiveness of breast cancer screening. A scoring scheme was used to score womens responses. A total of 900 women agreed to take part in the study. Their mean age was 41.6±12.4 years and 64% did not exceed the primary level of education. According to the constructed scores, 92% of participants had poor knowledge of the specific risk factors for breast cancer and 63.2% had poor knowledge of the screening methods. Proper practice of breast cancer screening was observed in 14.3% of cases. Multiple logistic regression analysis showed that good knowledge of risk factors and screening methods, higher level of education and positive family history of breast cancer were independently correlated with breast cancer screening practice. This study revealed poor knowledge of breast cancer and the screening methods as well as low levels of practice of breast cancer screening among women in the region of Monastir. Results justify educational programs to raise womens adherence to breast cancer screening programs in Tunisia.

Gender differences in adverse childhood experiences, collective violence, and the risk for addictive behaviors among university students in Tunisia

Adverse childhood experiences (ACE) have been linked to a variety of addictive behaviors. The recent adaptation of the ACE measure by the World Health Organization (WHO) allows for the assessment of the negative role of additional adverse experiences, such as extra-familial violence. To date, the relationship between extra-familial violence and addictive behaviors has not been assessed. We report the contribution of ACEs, including the new scales for extra-familial violence, on the risk for mental health problems and addictive behaviors by gender in a sample of young adults in Tunisia. We conducted a cross sectional study in Tunisia during 2014, where we recruited 1200 young university adults who completed the validated Arabic version of the WHO ACE questionnaire in a university setting. Results indicated that intra-familial adversities were associated with increased risk for addictive behaviors, particularly in males. ACEs were also associated with increased risk for mental health problems with women showing more difficulties than men. Exposure to peer, community and collective violence was higher in males than in females and logistic regression confirms that exposure to extra-familial violence increased the risk for addictive behaviors both in male and females by two to three-fold. Mental health problems were associated with peer violence and substance abuse in males, but not in females. Results demonstrate for the first time the contribution of exposure to extra-familial violence on risk for addictive behaviors. Results highlight the need for addressing mental health and addiction in a community with high burden of adversity and violence.

Predictors of smoking relapse in a cohort of adolescents and young adults in Monastir (Tunisia)

BackgroundSmoking prevalence in adolescents and young adults is substantially elevated in Tunisia. Moreover, there is a lack of knowledge regarding the effectiveness and associated factors in smoking cessation interventions among adolescents and young adults. This study aims at identifying the major factors leading to smoking relapse among adolescents and young adults in the region of Monastir, Tunisia.MethodsWe carried out a prospective cohort study at the smoking cessation center of the University hospital of Monastir, Tunisia. The population study consisted of all adolescents and young adults (15–30 years) consulted during a period of two years (2009 – 2010). A questionnaire was used to explore the patient’s sociodemographic characteristics, smoking history, nicotine dependence (Fagerstrom test) and anxiety / depression (Hospital Anxiety and Depression Scale). A telephone survey was conducted in July 2011 to assess smoking cessation results. A multivariate Cox regression was used to identify predictors of smoking relapses.ResultsA total of 221 adolescents and young adults were included in this study with a mean age of 25.5 ± 3.9 years. At follow up, 59 study participants (26.7%) were abstinent and the overall median abstinence was 2 months. In the multivariate analysis smoking relapse was associated with being an adolescent patient (HR 2.16; 95% CI: 1.54-3.05), medium or higher nicotine dependence at baseline (HR 2.66, 95% CI: 1.06-7.05 and HR 3.12, 95% CI: 1.20-8.12 respectively), not receiving treatment (HR 1.70, 95% CI: 1.25-2.33) and have friend who is a smoker (HR 1.63; 95% CI: 0.96-2.79).ConclusionThe results of this study provide important information about beneficial effect of smoking cessation support for adolescent and young adults. More efforts must be deployed to deal with contributing factors to smoking relapse.

Effectiveness of a tool for structuring action plan after analysis of adverse event

Context: Methods and tools which enabled healthcare organization for learning from adverse events focus mainly on reporting and analysis phases. However, formulating appropriate action plan is also a difficult phase that current methods do not help us with. Objective: To assess the effectiveness of a tool for structuring action plan for hospital risk managers. Method: A randomization procedure has been used to assign 56 voluntary hospital risk managers in two groups differing only by the use of the tool. Each group had to identify causes and formulate an action plan for two scenarios of adverse drug events successively. The first measure was realized without tool. In the second measure, only the intervention group used the tool. Results: The mean difference in the number of relevant actions proposed by participants between the two measures was statistically significant between the reference group (M = − 0.6, SD = 2.2) and the intervention group (M = 1.3, SD = 2.2), t(53) = 2.96, p = < .01). In a linear mixed effects model for longitudinal data, the number of proposed actions was statistically higher in the intervention group with a difference of 2.4 actions, 95% CI [1.43, 3.40], p < .001 and 1.1 actions, 95% CI [0.02, 2.00], p < .05) for scenario 1 and scenario 2 respectively. Discussion: Our study confirms the effectiveness of the tool in two scenarios. Further studies could assess its effectiveness in other adverse events, its feasibility and acceptability by hospital risk managers before proposing its broad use.

Rheumatic heart disease in a developing country: Incidence and trend (Monastir; Tunisia: 2000–2013)

BACKGROUND The penicillin therapy of β hemolytic streptococcal pharyngitis has aided in the decrease of rheumatic heart disease (RHD) in developing countries. Tunisia is an endemic area, however, and incidence of RHD is weakly documented. We aimed at establishing the standardized incidence rate (SIR) of RHD in Monastir governorate and at determining RHD prevalence among hospitalized patients in two cardiology departments. METHODS From the regional register of Monastir Hospital morbidity, we have selected newly diagnosed patients with RHD, residents of Monastir, and hospitalized to the 2 cardiology departments between 2000 and 2013 (2001 not included). FINDINGS We studied 676 newly admitted patients. We estimate 1060 to be the number of new annual RHD cases in Tunisia. The SIR per 105 person-years was 10.97, being 9.3 in men and 19.1 in women, respectively. We have notified a negative trend of crude incidence rate/105 Inhabitants (Inh) (CIR) (r=-0.23, p<10-3), and a strong positive correlation between age and CIR/105 Inh (r=0.989, p<10-4). RHD lethality was 1%. We have registered 728 hospitalizations for RHD, representing 2.5% of all cardiology hospitalizations [95% CI: 2.3-2.7%], with a prevalence for 13.3% for women aged 15-29years. The median hospital stay was 9days (IQR: 5-15). CONCLUSION Our results confirm the RHD incidence decrease, consistent with epidemiological transition in Tunisia. We have also emphasized on the close trend of RHD with age and the predominance of RHD among women especially at the procreation age.

Hospitalizations for communicable diseases in a developing country: prevalence and trends—Monastir, Tunisia, 2002–2013

BACKGROUND In spite of the epidemiological transition, communicable diseases remain a public health problem and represent a significant cause of morbidity and mortality worldwide, especially in developing countries. This study aimed to determine the crude and standardized prevalence rates of hospitalizations for communicable disease (HCD) and to assess trends in HCD by age and sex at a university hospital in Tunisia over a period of 12 years (2002-2013). METHODS All cases of HCD from 2002 to 2013 in the university hospital departments were included. Data collected from the regional register of hospital morbidity were used. The discharge diagnoses were coded according to the International Classification of Diseases, 10th revision (ICD-10). RESULTS HCD represented 17.45% of all hospitalizations during the study period (34 289/196 488; 95% confidence interval 17.28-17.62%). The median age at the time of admission was 31 years (interquartile range (IQR) 15-52 years). The median hospital length of stay (LOS) was 5days (IQR 3-9 days). The crude prevalence rate (CPR) was 5.41 per 1000 inhabitants. The CPR was highest among patients aged ≥65 years. The four communicable disease categories that represented 70% of all HCD were abdominal infection, skin infection, genitourinary infection, and lower respiratory tract infection. The majority of HCD decreased over time; however, there was a significant increase in HIV diseases, tuberculosis, and viral hepatitis. CONCLUSION This study provides evidence of the epidemiological transition, showing a decline in communicable diseases, which needs to be sustained and improved.

Childhood exposure to violence is associated with risk for mental disorders and adult’s weight status: a community-based study in Tunisia

Background We sought to investigate the relationship between social violence and adult overweight/obesity and the role of common mental disorders (CMD) in mediating this relationship. Methods A cross-sectional study was conducted from January to June 2016 in Tunisia. Participants were selected from randomly selected Primary Health Care Centers. The Arabic version of the Adverse Childhood Experiences-International Questionnaire (ACE-IQ) was used. Results A total of 2120 participants were included. Women exposed to social ACEs had higher rates of overweight/obesity than men (13.5 versus 9.5%; P = 0.004). For women, statistically significant partial mediation effects of CMD were observed for exposure to community violence (% mediated = 17.7%). For men, partial mediation was found for the exposure to peer violence (% mediated = 12.5%). Conclusion Our results provide evidence of the independent increase of overweight/obesity after exposure to social ACEs. Efforts to uncover and address underlying trauma in health care settings may increase the effectiveness of obesity interventions.

Prognostic value of hyperglycemia on-admission in diabetic versus non-diabetic patients presenting for ST-elevation myocardial infarction in Tunisia

Background Hyperglycemia on-admission is a powerful predictor of adverse events in patients presenting for ST-elevation myocardial infarction (STEMI). Aim In this study, we sought to determine the prognostic value of hyperglycemia on-admission in Tunisian patients presenting with STEMI according to their diabetic status. Methods Patients presenting to our center between January 1998 and September 2014 were enrolled. Hyperglycemia was defined as a glucose level ≥11 mmol/L. In-hospital prognosis was studied in diabetic and non-diabetic patients. The predictive value for mortality of glycemia level on-admission was assessed by mean of the area under receiver operating characteristic (ROC) curve calculation. Results A total of 1289 patients were included. Mean age was 60.39 ± 12.8 years and 977 (77.3%) patients were male. Prevalence of diabetes mellitus was 70.2% and 15.2% in patients presenting with and without hyperglycemia, respectively (p < 0.001). In univariate analysis, hyperglycemia was associated to in-hospital death in diabetic (OR: 8.85, 95% CI: 2.11–37.12, p < 0.001) and non-diabetic patients (OR: 2.57, 95% CI: 1.39–4.74, p = 0.002). In multivariate analysis, hyperglycemia was independently predictive of in-hospital death in diabetic patients (OR: 9.6, 95% CI: 2.18–42.22, p = 0.003) but not in non-diabetic patients (OR: 1.93, 95% CI: 0.97–3.86, p = 0.06). Area under ROC curve of glycemia as a predictor of in-hospital death was 0.792 in diabetic and 0.676 in non-diabetic patients. Conclusion In patients presenting with STEMI, hyperglycemia was associated to hospital death in diabetic and non-diabetic patients in univariate analysis. In multivariate analysis, hyperglycemia was independently associated to in-hospital death in diabetic but not in non-diabetic patients.