- Email: [email protected]
Obstetric competence among referral healthcare providers in Mali
IJG-07933; No of Pages 4 International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Obstetric competence among referral healthcare providers in Mali Cyrille Huchon a,b, Catherine Arsenault b, Caroline Tourigny b, Aliou Coulibaly b, Mamadou Traore c, Alexandre Dumont d, Pierre Fournier b,⁎ a
Centre de recherche du CHU Sainte Justine, Montréal, Canada Axe de Santé Mondiale, CRCHUM, Université de Montréal, Montréal, Canada URFOSAME, Faculté de Médecine, Université de Bamako, Bamako, Mali d Unité 216, Institut de recherche pour le développement, Paris, France b c
a r t i c l e
i n f o
Article history: Received 30 August 2013 Received in revised form 20 December 2013 Accepted 26 March 2014 Keywords: Clinical vignette Emergency obstetric care Obstetric knowledge Referral care Sub-Saharan Africa
a b s t r a c t Objective: To determine the factors associated with obstetric competency and clinical practice among obstetric care providers in referral health centers in Mali. Methods: The present cross-sectional study was conducted between March and May 2012 among 140 obstetric care providers (obstetric nurses, midwives, and physicians) working in referral health centers in Mali. Emergency obstetric care knowledge and skills were evaluated with clinical vignettes developed using national Malian guidelines. The vignettes covered 5 areas of emergency obstetric care, and the results were used to generate a competency score. A backward stepwise random-effects model using a maximum likelihood estimator was applied to evaluate variables independently associated with competency score. Results: Out of 100, the mean ± SD score was 57.8 ± 11.2 for obstetric nurses, 66.4 ± 14.7 for midwifes, and 78.6 ± 13.4 for physicians (P b 0.001). Three variables were significantly associated with a higher competency score: professional qualification, working in an urban setting, and working in a health center with a smaller number of obstetric care providers. Conclusion: Increasing the in-service training of both rural staff and lower-level healthcare workers working in larger health centers via facility-based maternal death reviews might help to improve clinical practice and maternal health outcomes. © 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction High rates of maternal mortality and morbidity persist in SubSaharan Africa, and progress in this area has been poor. Reducing these rates is the aim of the fifth Millennium Development Goal, which—as is now clear—will not be attained in this region of the world [1]. Although reducing maternal mortality and morbidity is feasible, it remains a major challenge in Sub-Saharan Africa, where healthcare systems are weak, and the availability and quality of care are heterogeneous and often inadequate [2–4]. The human resource crisis faced by many health systems in low-resource countries is characterized both by an insufficient number of staff and by a shortage of skills among existing healthcare workers. Furthermore, the competency of so-called “skilled birth attendants” has seldom been studied [5]. In Mali, the obstetric competence of healthcare providers at the community level can be deficient [6], increasing the risk of obstetric complications and delaying the timely referral of women to referral health centers, where comprehensive emergency obstetric and neonatal care ⁎ Corresponding author at: Axe de Santé Mondiale, Centre de Recherche du Centre Hospitalier de l’Université de Montréal (CRCHUM), 850 rue Saint-Denis, 3è étage, bureau S03-900, Montréal, Québec H2X 0A9, Canada. Tel.: +1 514 890 8000x15926. E-mail address: [email protected] (P. Fournier).
(EmONC) can be provided. Referral health centers have more advanced technologic platforms compared with community health centers (CHCs), and obstetric care therein is provided by personnel with specialized obstetric training including physicians with surgical skills (e.g. for cesarean delivery), midwives, obstetric nurses, and nurse anesthetists [7]. Obstetric nurses undergo 3 years of specialized training after primary school, midwives undergo 3 years of (different) specialized training after the Malian baccalaureate diploma, and physicians undergo 7 years of medical training after the baccalaureate diploma. Although the degree of specialization of healthcare providers can explain some of the variation in competency and clinical practice, assessment by title alone can result in gross mischaracterization [8]. Contextual factors such as the availability of equipment and supplies, the skills of surrounding coworkers, workload, access to clinical guidelines, feedback, and leadership might also influence a worker’s level of knowledge and skills. Knowing the influence of such institutional factors on competency and clinical practice might have the potential to improve maternal and neonatal outcomes by enhancing practice settings. The aim of the present study was to determine the factors associated with competency and clinical practice among obstetric care providers in referral health centers in 2 regions in Mali. In a separate study [6], the competency of primary healthcare workers has been assessed in CHCs in the same districts of Mali.
http://dx.doi.org/10.1016/j.ijgo.2014.01.014 0020-7292/© 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
2
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
2. Materials and methods
3. Results
The present cross-sectional survey was conducted among obstetric care workers in 9 referral health centers in 2 regions in Mali, the Kayes region and the capital Bamako, between March 1 and May 31, 2012. The study was approved by the ethics committees of the Research Centre of the University of Montreal Hospital, Montreal, Canada, and the Faculty of Medicine, Pharmacy and Odonto-Stomatology of the University of Bamako, Bamako, Mali. Informed consent was obtained from all participants. For the study, referral health centers were selected in districts corresponding to those of a previous survey at the CHC level [6]. Respondents were thus recruited in the Kayes regional hospital, in 7 “Centres de Santé de Référence” in the Kayes region, and in 1 “Centre de Santé de Référence” in the capital Bamako. In each center, all staff involved in maternal healthcare who had performed at least 1 delivery in the previous month were recruited. EmONC competency was evaluated via a test consisting of clinical vignettes (Supplementary Material S1), which were developed by Malian and French gynecologists and were based on the Malian national guidelines on obstetric care services [9]. The vignettes evaluated 5 areas of EmONC: acute complications during labor leading to hemorrhage and uterine rupture; hypertensive disorders (knowledge of pre-eclampsia and eclampsia); ability to supervise labor (dystocia); postpartum infection management; and neonatal care (Apgar score). Vignettes are known to be a valid and comprehensive method for measuring quality of clinical practice [10]. Some of the vignettes used in the present study were the same as those used in the Malian study in CHCs [6], but different answers were expected because healthcare workers in referral health centers must provide comprehensive EmONC as opposed to the basic EmONC provided in CHCs. Some vignettes described a more advanced stage of an obstetric complication included in the CHC study, reflecting the condition of women who come late to referral health centers. The same 3 basic knowledge questions used in the survey in CHCs were also included in the test: that is, naming the 5 criteria of the Apgar score, naming 3 signs of pre-eclampsia, and naming anticonvulsants in the case of eclampsia. A score was calculated on the basis of staff answers to the test. Between 1 and 5 items were expected for each answer, but a total of 1 point was given per question. The total score was converted into a percentage. The test was administrated orally by a Malian physician (A.C.) in French or Bambara. Institutional variables were collected via a questionnaire developed in a previous study in Senegal and Mali [11]. Data were collected on human resources, the size of the referral health center in terms of number of beds and number of staff, the availability of equipment, drugs, and supplies needed for EmONC, access to a medical library or the internet, and the availability of different comprehensive EmONC services. Data on the number of assisted deliveries performed per year in each referral health center were obtained from records held by the regional department of health. Statistical analyses were performed with Stata version 12.0 (StataCorp, College Station, TX, USA). After ensuring that no multicollinearity was present, bivariate analysis between the mean score and individual and institutional characteristics was explored via analysis of variance. All 12 individual and institutional independent variables were tested simultaneously in a random-effects model using a maximum likelihood estimator to evaluate variables associated with competency score, taking into account a potential cluster effect per health center. The final model was built in a backward stepwise fashion and included variables with a P value of less than 0.05. Interactions between variables were analyzed. Continuous variables were dichotomized on their median value for presentation in the univariate analysis, but were included as continuous variables in the randomeffects model.
All eligible healthcare workers agreed to participate, and 140 respondents from 9 referral health centers were recruited. The numbers of respondents in each health center ranged from 9 to 51. The distribution of respondents per referral health center and the individual years of experience of respondents are shown in Table 1. The mean total score for each area of assessment was plotted by type of healthcare worker (Fig. 1). The mean ± SD score was 66.3 ± 15.2 out of 100, and physicians scored highest (78.6 ± 13.4), followed by midwives (66.4 ± 14.7) and obstetric nurses (57.8 ± 11.2) (P b 0.001). This ranking was observed in each area of assessment except for questions related to neonatal care, where obstetric nurses and midwives scored slightly higher than physicians. All cadres combined, the healthcare workers scored highest on vignettes related to hypertensive complications (mean, 84.8), followed by Apgar score (mean, 79.1), hemorrhage and uterine ruptures (mean, 76.1), and dystocic labor (54.5), and scored lowest on vignettes related to postpartum infections (mean, 35.2). Univariate associations among all 12 independent variables and the competency score are presented in Table 2. The results of the final random-effects model of individual and institutional factors associated with competency score are shown in Table 3. In the final model, 3 variables remained significantly associated with a higher obstetric competency score: type of obstetric care provider, urban location, and total number of obstetric care providers at the health center (all P b 0.001). The latter variable was negatively associated with competency score, indicating that staff had a tendency to be more competent when working in a smaller referral health center. No interactions were significant and the cluster effect was negligible (likelihood-ratio test that the variance between centers = 0.00; P N 0.99). The predicted competency score, adjusted for other covariates in the model, was plotted by type of healthcare worker and location of the health center (Fig. 2). 4. Discussion The present study, based on an evaluation by clinical vignettes, found a need to improve the ability of healthcare workers to handle obstetric emergencies across referral health centers in Mali. A strength of the study is the use of clinical vignettes. Clinical vignettes are known to be valid and reliable tools for measuring clinical practice [10]. The present clinical vignettes were produced by using local guidelines for emergency obstetric care that seem more suitable for evaluating local practice, which differs from the context of highincome countries. Thus, the use of clinical vignettes derived from
Table 1 Characteristics of respondents.a Characteristic
Total individual experience, y Mean ± SD Median (range) Health centers Rural Diéma Bafoulabé Kéniéba Nioro Yélimané Urban Commune V, Bamako Kita Kayes-CSREF Kayes hospital a
Obstetric nurses (n = 43)
Midwives (n = 68)
Physicians (n = 29)
Total (n = 140)
9.4 ± 7.7 7 (1–33)
12.0 ± 9.9 8 (1–35)
7.5 ± 3.6 7 (1–15)
10.3 ± 8.4 7 (1–35)
14 (32.6) 5 (11.6) 2 (4.7) 3 (7.0) 1 (2.3) 3 (7.0) 29 (67.4) 15 (34.9) 8 (18.6) 5 (11.6) 1 (2.3)
17 (25.0) 3 (4.4) 4 (5.9) 3 (4.4) 5 (7.4) 2 (2.9) 51 (75.0) 32 (47.1) 6 (8.8) 6 (8.8) 7 (10.3)
17 (58.6) 4 (13.8) 4 (13.8) 4 (13.8) 2 (6.9) 3 (10.3) 12 (41.4) 4 (13.8) 4 (13.8) 2 (6.9) 2 (6.9)
48 (34.3) 12 (8.6) 10 (7.1) 10 (7.1) 8 (5.7) 8 (5.7) 92 (65.7) 51 (36.4) 18 (12.9) 13 (9.3) 10 (7.1)
Values are given as number (percentage) unless stated otherwise.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
3
Fig. 1. Competency score by area of assessment and type of healthcare provider in referral health centers in Mali.
Table 2 Univariate analyses of individual and institutional factors associated with obstetric competency score. Factor Individual characteristics Type of healthcare worker Obstetric nurses Midwives Physicians Total individual experience, y ≤7 N7 Institutional characteristics Location of center Rural Urban No. of obstetric care providers in center ≤20 N20 No. of assisted deliveries in 2011 ≤1500 N1500 Medical library or internet access No Yes Records kept for maternal deaths and referrals No Yes Specialized carea No Yes Obstetric ultrasound No Yes Blood transfusion and neonatal resuscitation No Yes Maternal cardiopulmonary resuscitation No Yes Number of beds in maternity ward ≤10 N10
Obstetric competency score, mean ± SD
P value
57.8 ± 11.2 66.4 ± 14.7 78.6 ± 13.4
b0.001
67.6 ± 14.9 64.9 ± 15.6
0.31
69.0 ± 1.6 62.8 ± 15.9
0.02
70.2 ± 14.3 62.8 ± 15.3
0.004
70.0 ± 14.3 62.9 ± 15.4
0.006
71.0 ± 15.3 63.9 ± 14.7
0.008
69.2 ± 13.7 63.3 ± 16.2
0.02
68.1 ± 14.3 64.9 ± 15.9
0.22
66.4 ± 10.2 66.3 ± 16.0
0.96
64.7 ± 11.8 66.6 ± 15.8
0.60
65.1 ± 11.2 66.7 ± 16.3
0.61
68.1 ± 14.3 64.9 ± 15.9
0.22
North American guidelines would have not been reliable for the present study. Another strength of the study is the exhaustive inclusion of all obstetric nurses, midwives, and physicians working in the different study centers, thereby avoiding selection bias. A limitation of the study is the inclusion of a health center with 51 respondents when the other 8 health centers had only between 8 and 18 obstetric care providers. However, this variation was addressed by using a random-effects model; in addition, when the analysis was repeated without the very large center, similar results were obtained (data not shown). In the multivariate model, type of healthcare worker was associated with obstetric competency score, and healthcare workers with a higher level of training scored higher in all areas except neonatal care. The Apgar score was better known by obstetric nurses and midwives than by physicians, which probably reflects the tendency of physicians to leave immediate newborn care to other staff. The multivariate model also showed that working in a rural referral health center is associated with lower obstetric competency regardless of professional qualification. This reflects disparities between urban and rural centers in terms of the amount of training and continuing education provided. Furthermore, urban centers might have an easier time recruiting more competent staff because retention of qualified personnel in rural regions is difficult [12]. Poorer neonatal care knowledge of staff from remote health centers was also found in northern Vietnam by Eriksson et al. [13]. In the study conducted in CHCs in Mali, however, rural location or distance to the referral health center was not associated with competency score [6]. The impact of geographic location on staff competency should be investigated further [14].
Table 3 Random-effects model of individual and contextual factors associated with obstetric competency score.a Variable a
a Including blood bank, neonatal intensive care unit, and pathologic pregnancy consultation.
Midwives Physiciansa Urban center Total no. of obstetric care providers in center
Coefficient
P value
8.82 18.2 15.7 –0.32
b0.001 b0.001 0.004 b0.001
Likelihood ratio (χ2) test statistic (4 degrees of freedom), 57.87; P b 0.001. a Reference, obstetric nurses.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
4
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Acknowledgments This research was supported by the Teasdale-Corti Global Health Research Partnership Program of the Global Health Research Initiative (Canadian Institutes of Health Research, Canadian International Development Agency, Health Canada, International Development Research Center and Public Health Agency of Canada). C.H. was supported by the Strategic Training Programin Global Health Research, a partnership of the Canadian Institutes of Health Research and the Québec Population Health Research Network. Conflict of interest The authors have no conflicts of interest. References
Fig. 2. Predictive margins of competency score by type of healthcare provider and location of health center with 95% confidence intervals.
The present study also found a negative association between size of health center in terms of total staff number and competency score. Although the interaction was not significant, this association seemed more pronounced for obstetric nurses and midwives. Obstetric nurses and midwives working in a health center with a higher number of obstetric care providers seemed to have poorer EmONC skills. This might be because in larger health centers, where more physicians are available, care of complicated deliveries is transferred directly to physicians, and obstetric nurses and midwives perform mostly non-complicated deliveries and routine care. Being surrounded by more physicians might be expected to have the effect of increasing the competency of lower-level staff; however, the opposite association was found in the present study. By not having the opportunity to handle obstetric emergencies, midwives and obstetric nurses in larger centers may not gain the experience required to build and maintain these skills, resulting in the disengagement of these cadres in emergency obstetric care. It has been argued that a certain level of activity is required in order to maintain skills [15,16]. In the present study, however, the number of deliveries performed per year at the health center was not retained in the final model. These conflicting results might simply be due to a lack of statistical power, but they are consistent with another study that found that the relationship between caseload and skills is unclear [17]. Although the delegation of tasks among physicians, midwives, and obstetric nurses is understandable, strengthening the EmONC skills of lower-level personnel is necessary. Organizing meetings involving all levels of obstetric care providers to discuss past cases of obstetric complications might help alleviate this perverse effect in health centers with large staff numbers. The use of facility-based maternal death reviews has been shown to be an efficient way to decrease avoidable maternal deaths. The ALARM program—a continuing education program developed by the society of obstetricians and gynecologists of Canada—had the greatest impact in rural health centers across Senegal and Mali, showing that there is potential for improving maternal healthcare in rural centers [11]. Increasing in-service training in rural health centers should be a priority.
[1] Salama P, Lawn J, Bryce J, Bustreo F, Fauveau V, Starrs A, et al. Making the Countdown count. Lancet 2008;371(9620):1219–21. [2] Bailey PE, Paxton A. Program note. Using UN process indicators to assess needs in emergency obstetric services. Int J Gynecol Obstet 2002;76(3):299–306. [3] Fauveau V. UNFPA–AMDD Working Group on Indicators. Program Note: Using UN process indicators to assess needs in emergency obstetric services: Gabon, Guinea-Bissau, and The Gambia. Int J Gynecol Obstet 2007;96(3):233–40. [4] Otchere SA, Binh HT. Strengthening emergency obstetric care in Thanh Hoa and Quang Tri provinces in Vietnam. Int J Gynecol Obstet 2007;99(2):165–72. [5] Harvey SA, Blandón YC, McCaw-Binns A, Sandino I, Urbina L, Rodríguez C, et al. Are skilled birth attendants really skilled? A measurement method, some disturbing results and a potential way forward. Bull World Health Organ 2007;85(10):783–90. [6] Traoré M, Arsenault C, Schoemaker-Marcotte C, Coulibaly A, Huchon C, Dumont A, et al. Obstetric competence among primary healthcare workers in Mali. Int J Gynecol Obstet 2014 (in press). [7] Dogba M, Fournier P, Dumont A, Zunzunegui MV, Tourigny C, Berthe-Cisse S. Mother and newborn survival according to point of entry and type of human resources in a maternal referral system in Kayes (Mali). Reprod Health 2011;8:13. [8] Spangler SA. Assessing skilled birth attendants and emergency obstetric care in rural Tanzania: the inadequacy of using global standards and indicators to measure local realities. Reprod Health Matters 2012;20(39):133–41. [9] Ministry of Health of Mali, Policies and procedures for and Reproductive Health. Volume 4: Women's Health: Prenatal, intrapartum and postpartum care, post-abortion care, gynecological disorders and sexual dysfunctions in women. http://www.sante.gov.ml/ATN1/Materials/repro_health/PNP6_Volume4.pdf. Published 2005. [10] Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients, and chart abstraction: a prospective validation study of 3 methods for measuring quality. JAMA 2000;283(13):1715–22. [11] Dumont A, Fournier P, Abrahamowicz M, Traoré M, Haddad S, Fraser WD, et al. Quality of care, risk management, and technology in obstetrics to reduce hospitalbased maternal mortality in Senegal and Mali (QUARITE): a cluster-randomised trial. Lancet 2013;382(9887):146–57. [12] Dussault G, Franceschini MC. Not enough there, too many here: understanding geographical imbalances in the distribution of the health workforce. Hum Resour Health 2006;4:12. [13] Eriksson L, Nga NT, Målqvist M, Persson LA, Ewald U, Wallin L. Evidence-based practice in neonatal health: knowledge among primary health care staff in northern Viet Nam. Hum Resour Health 2009;7:36. [14] Andrews GJ, Moon G. Space, place, and the evidence base: Part II–Rereading nursing environment through geographical research. Worldviews Evid Based Nurs 2005;2(3):142–56. [15] Keirse MJNC. Interaction between primary and secondary care during pregnancy and childbirth. In: Chalmers I, Enkin M, Keirse MJNC, editors. Effective Care in Pregnancy and Childbirth: Pregnancy, vol. 1. Oxford: Oxford Medical Publications; 1989. p. 197–201. [16] Scotland GS, Bullough CH. What do doctors think their caseload should be to maintain their skills for delivery care? Int J Gynecol Obstet 2004;87(3):301–7. [17] Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002;137(6): 511–20.
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijgo.2014.01.014.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Obstetric competence among referral healthcare providers in Mali Cyrille Huchon a,b, Catherine Arsenault b, Caroline Tourigny b, Aliou Coulibaly b, Mamadou Traore c, Alexandre Dumont d, Pierre Fournier b,⁎ a
Centre de recherche du CHU Sainte Justine, Montréal, Canada Axe de Santé Mondiale, CRCHUM, Université de Montréal, Montréal, Canada URFOSAME, Faculté de Médecine, Université de Bamako, Bamako, Mali d Unité 216, Institut de recherche pour le développement, Paris, France b c
a r t i c l e
i n f o
Article history: Received 30 August 2013 Received in revised form 20 December 2013 Accepted 26 March 2014 Keywords: Clinical vignette Emergency obstetric care Obstetric knowledge Referral care Sub-Saharan Africa
a b s t r a c t Objective: To determine the factors associated with obstetric competency and clinical practice among obstetric care providers in referral health centers in Mali. Methods: The present cross-sectional study was conducted between March and May 2012 among 140 obstetric care providers (obstetric nurses, midwives, and physicians) working in referral health centers in Mali. Emergency obstetric care knowledge and skills were evaluated with clinical vignettes developed using national Malian guidelines. The vignettes covered 5 areas of emergency obstetric care, and the results were used to generate a competency score. A backward stepwise random-effects model using a maximum likelihood estimator was applied to evaluate variables independently associated with competency score. Results: Out of 100, the mean ± SD score was 57.8 ± 11.2 for obstetric nurses, 66.4 ± 14.7 for midwifes, and 78.6 ± 13.4 for physicians (P b 0.001). Three variables were significantly associated with a higher competency score: professional qualification, working in an urban setting, and working in a health center with a smaller number of obstetric care providers. Conclusion: Increasing the in-service training of both rural staff and lower-level healthcare workers working in larger health centers via facility-based maternal death reviews might help to improve clinical practice and maternal health outcomes. © 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction High rates of maternal mortality and morbidity persist in SubSaharan Africa, and progress in this area has been poor. Reducing these rates is the aim of the fifth Millennium Development Goal, which—as is now clear—will not be attained in this region of the world [1]. Although reducing maternal mortality and morbidity is feasible, it remains a major challenge in Sub-Saharan Africa, where healthcare systems are weak, and the availability and quality of care are heterogeneous and often inadequate [2–4]. The human resource crisis faced by many health systems in low-resource countries is characterized both by an insufficient number of staff and by a shortage of skills among existing healthcare workers. Furthermore, the competency of so-called “skilled birth attendants” has seldom been studied [5]. In Mali, the obstetric competence of healthcare providers at the community level can be deficient [6], increasing the risk of obstetric complications and delaying the timely referral of women to referral health centers, where comprehensive emergency obstetric and neonatal care ⁎ Corresponding author at: Axe de Santé Mondiale, Centre de Recherche du Centre Hospitalier de l’Université de Montréal (CRCHUM), 850 rue Saint-Denis, 3è étage, bureau S03-900, Montréal, Québec H2X 0A9, Canada. Tel.: +1 514 890 8000x15926. E-mail address: [email protected] (P. Fournier).
(EmONC) can be provided. Referral health centers have more advanced technologic platforms compared with community health centers (CHCs), and obstetric care therein is provided by personnel with specialized obstetric training including physicians with surgical skills (e.g. for cesarean delivery), midwives, obstetric nurses, and nurse anesthetists [7]. Obstetric nurses undergo 3 years of specialized training after primary school, midwives undergo 3 years of (different) specialized training after the Malian baccalaureate diploma, and physicians undergo 7 years of medical training after the baccalaureate diploma. Although the degree of specialization of healthcare providers can explain some of the variation in competency and clinical practice, assessment by title alone can result in gross mischaracterization [8]. Contextual factors such as the availability of equipment and supplies, the skills of surrounding coworkers, workload, access to clinical guidelines, feedback, and leadership might also influence a worker’s level of knowledge and skills. Knowing the influence of such institutional factors on competency and clinical practice might have the potential to improve maternal and neonatal outcomes by enhancing practice settings. The aim of the present study was to determine the factors associated with competency and clinical practice among obstetric care providers in referral health centers in 2 regions in Mali. In a separate study [6], the competency of primary healthcare workers has been assessed in CHCs in the same districts of Mali.
http://dx.doi.org/10.1016/j.ijgo.2014.01.014 0020-7292/© 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
2
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
2. Materials and methods
3. Results
The present cross-sectional survey was conducted among obstetric care workers in 9 referral health centers in 2 regions in Mali, the Kayes region and the capital Bamako, between March 1 and May 31, 2012. The study was approved by the ethics committees of the Research Centre of the University of Montreal Hospital, Montreal, Canada, and the Faculty of Medicine, Pharmacy and Odonto-Stomatology of the University of Bamako, Bamako, Mali. Informed consent was obtained from all participants. For the study, referral health centers were selected in districts corresponding to those of a previous survey at the CHC level [6]. Respondents were thus recruited in the Kayes regional hospital, in 7 “Centres de Santé de Référence” in the Kayes region, and in 1 “Centre de Santé de Référence” in the capital Bamako. In each center, all staff involved in maternal healthcare who had performed at least 1 delivery in the previous month were recruited. EmONC competency was evaluated via a test consisting of clinical vignettes (Supplementary Material S1), which were developed by Malian and French gynecologists and were based on the Malian national guidelines on obstetric care services [9]. The vignettes evaluated 5 areas of EmONC: acute complications during labor leading to hemorrhage and uterine rupture; hypertensive disorders (knowledge of pre-eclampsia and eclampsia); ability to supervise labor (dystocia); postpartum infection management; and neonatal care (Apgar score). Vignettes are known to be a valid and comprehensive method for measuring quality of clinical practice [10]. Some of the vignettes used in the present study were the same as those used in the Malian study in CHCs [6], but different answers were expected because healthcare workers in referral health centers must provide comprehensive EmONC as opposed to the basic EmONC provided in CHCs. Some vignettes described a more advanced stage of an obstetric complication included in the CHC study, reflecting the condition of women who come late to referral health centers. The same 3 basic knowledge questions used in the survey in CHCs were also included in the test: that is, naming the 5 criteria of the Apgar score, naming 3 signs of pre-eclampsia, and naming anticonvulsants in the case of eclampsia. A score was calculated on the basis of staff answers to the test. Between 1 and 5 items were expected for each answer, but a total of 1 point was given per question. The total score was converted into a percentage. The test was administrated orally by a Malian physician (A.C.) in French or Bambara. Institutional variables were collected via a questionnaire developed in a previous study in Senegal and Mali [11]. Data were collected on human resources, the size of the referral health center in terms of number of beds and number of staff, the availability of equipment, drugs, and supplies needed for EmONC, access to a medical library or the internet, and the availability of different comprehensive EmONC services. Data on the number of assisted deliveries performed per year in each referral health center were obtained from records held by the regional department of health. Statistical analyses were performed with Stata version 12.0 (StataCorp, College Station, TX, USA). After ensuring that no multicollinearity was present, bivariate analysis between the mean score and individual and institutional characteristics was explored via analysis of variance. All 12 individual and institutional independent variables were tested simultaneously in a random-effects model using a maximum likelihood estimator to evaluate variables associated with competency score, taking into account a potential cluster effect per health center. The final model was built in a backward stepwise fashion and included variables with a P value of less than 0.05. Interactions between variables were analyzed. Continuous variables were dichotomized on their median value for presentation in the univariate analysis, but were included as continuous variables in the randomeffects model.
All eligible healthcare workers agreed to participate, and 140 respondents from 9 referral health centers were recruited. The numbers of respondents in each health center ranged from 9 to 51. The distribution of respondents per referral health center and the individual years of experience of respondents are shown in Table 1. The mean total score for each area of assessment was plotted by type of healthcare worker (Fig. 1). The mean ± SD score was 66.3 ± 15.2 out of 100, and physicians scored highest (78.6 ± 13.4), followed by midwives (66.4 ± 14.7) and obstetric nurses (57.8 ± 11.2) (P b 0.001). This ranking was observed in each area of assessment except for questions related to neonatal care, where obstetric nurses and midwives scored slightly higher than physicians. All cadres combined, the healthcare workers scored highest on vignettes related to hypertensive complications (mean, 84.8), followed by Apgar score (mean, 79.1), hemorrhage and uterine ruptures (mean, 76.1), and dystocic labor (54.5), and scored lowest on vignettes related to postpartum infections (mean, 35.2). Univariate associations among all 12 independent variables and the competency score are presented in Table 2. The results of the final random-effects model of individual and institutional factors associated with competency score are shown in Table 3. In the final model, 3 variables remained significantly associated with a higher obstetric competency score: type of obstetric care provider, urban location, and total number of obstetric care providers at the health center (all P b 0.001). The latter variable was negatively associated with competency score, indicating that staff had a tendency to be more competent when working in a smaller referral health center. No interactions were significant and the cluster effect was negligible (likelihood-ratio test that the variance between centers = 0.00; P N 0.99). The predicted competency score, adjusted for other covariates in the model, was plotted by type of healthcare worker and location of the health center (Fig. 2). 4. Discussion The present study, based on an evaluation by clinical vignettes, found a need to improve the ability of healthcare workers to handle obstetric emergencies across referral health centers in Mali. A strength of the study is the use of clinical vignettes. Clinical vignettes are known to be valid and reliable tools for measuring clinical practice [10]. The present clinical vignettes were produced by using local guidelines for emergency obstetric care that seem more suitable for evaluating local practice, which differs from the context of highincome countries. Thus, the use of clinical vignettes derived from
Table 1 Characteristics of respondents.a Characteristic
Total individual experience, y Mean ± SD Median (range) Health centers Rural Diéma Bafoulabé Kéniéba Nioro Yélimané Urban Commune V, Bamako Kita Kayes-CSREF Kayes hospital a
Obstetric nurses (n = 43)
Midwives (n = 68)
Physicians (n = 29)
Total (n = 140)
9.4 ± 7.7 7 (1–33)
12.0 ± 9.9 8 (1–35)
7.5 ± 3.6 7 (1–15)
10.3 ± 8.4 7 (1–35)
14 (32.6) 5 (11.6) 2 (4.7) 3 (7.0) 1 (2.3) 3 (7.0) 29 (67.4) 15 (34.9) 8 (18.6) 5 (11.6) 1 (2.3)
17 (25.0) 3 (4.4) 4 (5.9) 3 (4.4) 5 (7.4) 2 (2.9) 51 (75.0) 32 (47.1) 6 (8.8) 6 (8.8) 7 (10.3)
17 (58.6) 4 (13.8) 4 (13.8) 4 (13.8) 2 (6.9) 3 (10.3) 12 (41.4) 4 (13.8) 4 (13.8) 2 (6.9) 2 (6.9)
48 (34.3) 12 (8.6) 10 (7.1) 10 (7.1) 8 (5.7) 8 (5.7) 92 (65.7) 51 (36.4) 18 (12.9) 13 (9.3) 10 (7.1)
Values are given as number (percentage) unless stated otherwise.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
3
Fig. 1. Competency score by area of assessment and type of healthcare provider in referral health centers in Mali.
Table 2 Univariate analyses of individual and institutional factors associated with obstetric competency score. Factor Individual characteristics Type of healthcare worker Obstetric nurses Midwives Physicians Total individual experience, y ≤7 N7 Institutional characteristics Location of center Rural Urban No. of obstetric care providers in center ≤20 N20 No. of assisted deliveries in 2011 ≤1500 N1500 Medical library or internet access No Yes Records kept for maternal deaths and referrals No Yes Specialized carea No Yes Obstetric ultrasound No Yes Blood transfusion and neonatal resuscitation No Yes Maternal cardiopulmonary resuscitation No Yes Number of beds in maternity ward ≤10 N10
Obstetric competency score, mean ± SD
P value
57.8 ± 11.2 66.4 ± 14.7 78.6 ± 13.4
b0.001
67.6 ± 14.9 64.9 ± 15.6
0.31
69.0 ± 1.6 62.8 ± 15.9
0.02
70.2 ± 14.3 62.8 ± 15.3
0.004
70.0 ± 14.3 62.9 ± 15.4
0.006
71.0 ± 15.3 63.9 ± 14.7
0.008
69.2 ± 13.7 63.3 ± 16.2
0.02
68.1 ± 14.3 64.9 ± 15.9
0.22
66.4 ± 10.2 66.3 ± 16.0
0.96
64.7 ± 11.8 66.6 ± 15.8
0.60
65.1 ± 11.2 66.7 ± 16.3
0.61
68.1 ± 14.3 64.9 ± 15.9
0.22
North American guidelines would have not been reliable for the present study. Another strength of the study is the exhaustive inclusion of all obstetric nurses, midwives, and physicians working in the different study centers, thereby avoiding selection bias. A limitation of the study is the inclusion of a health center with 51 respondents when the other 8 health centers had only between 8 and 18 obstetric care providers. However, this variation was addressed by using a random-effects model; in addition, when the analysis was repeated without the very large center, similar results were obtained (data not shown). In the multivariate model, type of healthcare worker was associated with obstetric competency score, and healthcare workers with a higher level of training scored higher in all areas except neonatal care. The Apgar score was better known by obstetric nurses and midwives than by physicians, which probably reflects the tendency of physicians to leave immediate newborn care to other staff. The multivariate model also showed that working in a rural referral health center is associated with lower obstetric competency regardless of professional qualification. This reflects disparities between urban and rural centers in terms of the amount of training and continuing education provided. Furthermore, urban centers might have an easier time recruiting more competent staff because retention of qualified personnel in rural regions is difficult [12]. Poorer neonatal care knowledge of staff from remote health centers was also found in northern Vietnam by Eriksson et al. [13]. In the study conducted in CHCs in Mali, however, rural location or distance to the referral health center was not associated with competency score [6]. The impact of geographic location on staff competency should be investigated further [14].
Table 3 Random-effects model of individual and contextual factors associated with obstetric competency score.a Variable a
a Including blood bank, neonatal intensive care unit, and pathologic pregnancy consultation.
Midwives Physiciansa Urban center Total no. of obstetric care providers in center
Coefficient
P value
8.82 18.2 15.7 –0.32
b0.001 b0.001 0.004 b0.001
Likelihood ratio (χ2) test statistic (4 degrees of freedom), 57.87; P b 0.001. a Reference, obstetric nurses.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014
4
C. Huchon et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Acknowledgments This research was supported by the Teasdale-Corti Global Health Research Partnership Program of the Global Health Research Initiative (Canadian Institutes of Health Research, Canadian International Development Agency, Health Canada, International Development Research Center and Public Health Agency of Canada). C.H. was supported by the Strategic Training Programin Global Health Research, a partnership of the Canadian Institutes of Health Research and the Québec Population Health Research Network. Conflict of interest The authors have no conflicts of interest. References
Fig. 2. Predictive margins of competency score by type of healthcare provider and location of health center with 95% confidence intervals.
The present study also found a negative association between size of health center in terms of total staff number and competency score. Although the interaction was not significant, this association seemed more pronounced for obstetric nurses and midwives. Obstetric nurses and midwives working in a health center with a higher number of obstetric care providers seemed to have poorer EmONC skills. This might be because in larger health centers, where more physicians are available, care of complicated deliveries is transferred directly to physicians, and obstetric nurses and midwives perform mostly non-complicated deliveries and routine care. Being surrounded by more physicians might be expected to have the effect of increasing the competency of lower-level staff; however, the opposite association was found in the present study. By not having the opportunity to handle obstetric emergencies, midwives and obstetric nurses in larger centers may not gain the experience required to build and maintain these skills, resulting in the disengagement of these cadres in emergency obstetric care. It has been argued that a certain level of activity is required in order to maintain skills [15,16]. In the present study, however, the number of deliveries performed per year at the health center was not retained in the final model. These conflicting results might simply be due to a lack of statistical power, but they are consistent with another study that found that the relationship between caseload and skills is unclear [17]. Although the delegation of tasks among physicians, midwives, and obstetric nurses is understandable, strengthening the EmONC skills of lower-level personnel is necessary. Organizing meetings involving all levels of obstetric care providers to discuss past cases of obstetric complications might help alleviate this perverse effect in health centers with large staff numbers. The use of facility-based maternal death reviews has been shown to be an efficient way to decrease avoidable maternal deaths. The ALARM program—a continuing education program developed by the society of obstetricians and gynecologists of Canada—had the greatest impact in rural health centers across Senegal and Mali, showing that there is potential for improving maternal healthcare in rural centers [11]. Increasing in-service training in rural health centers should be a priority.
[1] Salama P, Lawn J, Bryce J, Bustreo F, Fauveau V, Starrs A, et al. Making the Countdown count. Lancet 2008;371(9620):1219–21. [2] Bailey PE, Paxton A. Program note. Using UN process indicators to assess needs in emergency obstetric services. Int J Gynecol Obstet 2002;76(3):299–306. [3] Fauveau V. UNFPA–AMDD Working Group on Indicators. Program Note: Using UN process indicators to assess needs in emergency obstetric services: Gabon, Guinea-Bissau, and The Gambia. Int J Gynecol Obstet 2007;96(3):233–40. [4] Otchere SA, Binh HT. Strengthening emergency obstetric care in Thanh Hoa and Quang Tri provinces in Vietnam. Int J Gynecol Obstet 2007;99(2):165–72. [5] Harvey SA, Blandón YC, McCaw-Binns A, Sandino I, Urbina L, Rodríguez C, et al. Are skilled birth attendants really skilled? A measurement method, some disturbing results and a potential way forward. Bull World Health Organ 2007;85(10):783–90. [6] Traoré M, Arsenault C, Schoemaker-Marcotte C, Coulibaly A, Huchon C, Dumont A, et al. Obstetric competence among primary healthcare workers in Mali. Int J Gynecol Obstet 2014 (in press). [7] Dogba M, Fournier P, Dumont A, Zunzunegui MV, Tourigny C, Berthe-Cisse S. Mother and newborn survival according to point of entry and type of human resources in a maternal referral system in Kayes (Mali). Reprod Health 2011;8:13. [8] Spangler SA. Assessing skilled birth attendants and emergency obstetric care in rural Tanzania: the inadequacy of using global standards and indicators to measure local realities. Reprod Health Matters 2012;20(39):133–41. [9] Ministry of Health of Mali, Policies and procedures for and Reproductive Health. Volume 4: Women's Health: Prenatal, intrapartum and postpartum care, post-abortion care, gynecological disorders and sexual dysfunctions in women. http://www.sante.gov.ml/ATN1/Materials/repro_health/PNP6_Volume4.pdf. Published 2005. [10] Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients, and chart abstraction: a prospective validation study of 3 methods for measuring quality. JAMA 2000;283(13):1715–22. [11] Dumont A, Fournier P, Abrahamowicz M, Traoré M, Haddad S, Fraser WD, et al. Quality of care, risk management, and technology in obstetrics to reduce hospitalbased maternal mortality in Senegal and Mali (QUARITE): a cluster-randomised trial. Lancet 2013;382(9887):146–57. [12] Dussault G, Franceschini MC. Not enough there, too many here: understanding geographical imbalances in the distribution of the health workforce. Hum Resour Health 2006;4:12. [13] Eriksson L, Nga NT, Målqvist M, Persson LA, Ewald U, Wallin L. Evidence-based practice in neonatal health: knowledge among primary health care staff in northern Viet Nam. Hum Resour Health 2009;7:36. [14] Andrews GJ, Moon G. Space, place, and the evidence base: Part II–Rereading nursing environment through geographical research. Worldviews Evid Based Nurs 2005;2(3):142–56. [15] Keirse MJNC. Interaction between primary and secondary care during pregnancy and childbirth. In: Chalmers I, Enkin M, Keirse MJNC, editors. Effective Care in Pregnancy and Childbirth: Pregnancy, vol. 1. Oxford: Oxford Medical Publications; 1989. p. 197–201. [16] Scotland GS, Bullough CH. What do doctors think their caseload should be to maintain their skills for delivery care? Int J Gynecol Obstet 2004;87(3):301–7. [17] Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002;137(6): 511–20.
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijgo.2014.01.014.
Please cite this article as: Huchon C, et al, Obstetric competence among referral healthcare providers in Mali, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.01.014