Selection of health areas: the method used was non-probabilistic and purposive sampling. A total of ten (10) health areas out of twenty-one (21) were selected. In each health area, one (01) community health center was selected. In addition to these first-level community contact health structures, the district reference health center was also chosen. These health facilities were selected according to the following criteria:

Selection of mothers or caregivers of malnourished children aged 6 - 59 months: the probability sampling method was used through simple random sampling in the selected health areas.

The sample size was calculated according to the following formula:

$n=\frac{Z^{2}PQ}{I^2}=\frac{{1.96}^2\times 0.089\times \left(1-0.089\right)}{{0.05}^2}=125$

n: Size = 125; Z: Accuracy = 1.96.

P: Prevalence = 8.9% according to SMART 2018; I: Risk = 5%.

For the sample to be representative, we added 10% of the calculated sample size. A total of 138 mothers or carers of malnourished children aged 6 - 59 months were surveyed, i.e. 12 people per health area and 18 at the URENI of the reference health center.

Selection of health personnel: ten (10) technical directors of community health centers (TDCHC) out of twenty-one (21) and the head of the URENI of the health district’s reference health center were selected. A total of 149 subjects participated in the study.

Data collection procedure and tools used:

The literature review was carried out using a tabulation sheet. Once verbal informed consent had been granted by the local and village authorities, we asked for a community liaison officer to guide us in administering the questionnaire to the participants. The questionnaire was used to collect data from 20 participants during the pilot phase of the study and then validated by the research team. Under the guidance of three supervisors, six trained interviewers visited the selected facilities to collect data using questionnaires designed for this purpose. In addition to the questionnaires, the other tools used were:

Data collected: the information collected included.

Once the data had been processed, we corrected them to ensure completeness, consistency and concordance.

The data collected was entered into Microsoft Excel and then analyzed using Epi-Info version 7 software.

Given the purpose of the study, data analysis was limited to a descriptive analysis. All variables were described in percentage terms.

The purpose of the study was explained to the district’s political and administrative authorities and to the participants in order to obtain their consent.

Verbal consent was obtained from all participants. Confidentiality and anonymity were respected throughout the study process.

Description of integrated management of acute malnutrition in Mali ( Figure 1 )

According to the IMAM (Integrated Management of Acute Malnutrition) approach in Mali, there are three malnutrition management units:

Principles of care for acute malnutrition and operational definitions

The principles for management of severe acute malnutrition (SAM), whatever the program type includes three phases:

Exit: is defined as a patient leaving a care unit (URENI/URENAS/URENAM) or completely leaving the PCIMA program. This is the sum of patients; cured/successfully treated, deceased, dropped out, referred (including medically referred cases) and transferred.

Cured: is defined as a patient achieving the following cure criteria:

Child admitted with W/H (Weight/height) < −3 z-score only (A1): Achieved W/H (Weight/height) ≥ −2 and no edema for two consecutive visits*;

Child admitted with AP (arm perimeter) < 115 mm only (A2): Achieved AP ≥ 125 mm without edema during two consecutive visits;

Child admitted with W/H < −3 z-score and AP < 115 mm (A3): Achieved AP ≥ 125 mm and W/H ≥ −2 and no edema for two consecutive visits.

Deceased or Death: is defined as a patient who dies during their stay in the URENI/URENAS/URENAM program after registration. NB: This includes patients who die during transfer from one health facility to another. If the SAM patient dies during referral from URENAS to URENI, the death must be counted as a death in the program and noted in the URENAS report.

Abandonment: is defined as a patient absent for two consecutive weighing (in URENAS, 2 appointments at URENAM and 2 days in URENI).

Among the mothers of children who took part in the study, 85.51% (n = 118) were housewives, 87.68% (n = 121) were married, 3.68 (n = 5) were divorced and 76.09% (n = 105) were not educated.

Among malnourished children, the female sex, the 12 - 23 months age group and the severe form of malnutrition were the most represented, with 51.45%, 54.34% and 69.57% respectively. Only 48.55% of these children were fully vaccinated.

Concerning the characteristics of the health personnel surveyed, 100% (n = 11) were male, 81.82% were doctors, 18.18 (n = 2) were nurses and 90.90% (n = 10) had less than one year’s professional experience. Of these health professionals, 90.90% (n = 10) had received training in integrated management of acute malnutrition (IMPAM) and only 01 (9.09%) in policy, standards and procedures (PNP).

Breast-feeding was observed in 63% of malnourished children, and more than half the mothers (58.62%) gave ready-to-use therapeutic food (RUTF) before their children breastfeeds.

Among breastfed children, 64.37% benefited from breastfeeding frequency in line with standards (greater than or equal to eight times a day).

Thirty-seven percent (37%) of children had already been weaned prior to admission, and 82.35% of them had undergone abrupt weaning. Pregnancy was the main cause of weaning (68.63%).

This study revealed that anthropometric equipment, the national protocol for the management of acute malnutrition and nutritional inputs were available and in good condition. It also revealed that two centers out of ten (18.18%) did not have functional drinking water points. With regard to governance, the centers

evaluated had notes on the distribution of tasks between staff and a schedule for organizing nutrition education sessions and management committee coordination meetings ( Table 3 ).

The number of malnourished people admitted to the program was significantly higher than expected in seven out of ten facilities. The highest admission rate was observed at the Debere community health center (352.42%), and the lowest at the Bore CHC (24.62%). Rates exceeding one hundred percent can be explained by the fact that patients from outside the health areas concerned were also admitted to the program for various reasons, notably those linked to the internal displacement of populations due to insecurity ( Figure 2 ).

Cure rates were 81% in URENAM, 84% in URENAS and 92% in URENI. Drop-out rates were 19% in URENAM, 16% in URENAS and 0% in URENI. The death rate was 8% in URENI and 0% in URENAM and URENAS respectively ( Table 4 ).

URENAM*: Moderate outpatient nutritional recovery and education unit, URENAS*: Severe ambulatory nutritional recovery and education unit, URENI*: Intensive nutritional recovery and education unit.

Among the women questioned, only 61.59% said they had been informed of their children’s diagnosis, 84.78% had received nutritional advice, 63.77% knew the danger signs, and over 95% were assiduous in the program. More than half the mothers of malnourished children aged 6 - 59 months (67.39%) had a good assessment of IMPAM services ( Table 5 ).

All of the health workers (100%) surveyed affirmed that the community conflicts have a negative influence on integrated management program for acute malnutrition. According to them, the difficulties that hindered the correct implementation of the program were among others:

In our study, as in that conducted by I. Nantoume [6] , females predominated, in contrast to several other studies in the literature [7] - [9] . From a sociological point of view, this predominance of females is a major advantage in terms of the quality of care provided to children and the quality of nutritional advice given.

The 12 - 23 months age group was the most represented with 54.34%. This finding has also been reported in several literature series [3] [6] [10] [11] , and may be due to the fact that this age group is the most affected by poor complementary feeding, early weaning or ablactation, due to mothers’ lack of knowledge of good nutritional practice.

According to the form of acute malnutrition, 69.57% of our sample suffered from severe acute malnutrition. The dominance of the severe form of malnutrition could be explained by the late arrival of children in care. Added to this is their non-vaccination status (21%). This state of affairs has an influence on children’s nutritional status, as the weaker the immune system, the more exposed it is to disease, which is one of the immediate causes of malnutrition.

As for the mothers of the children included in the study, most of them had no formal education, as some authors have also reported [8] [12] . This state of affairs has a negative impact on the application of good nutritional practice methods, as well as on child weaning. Women’s lack of knowledge or education about family planning is one of the factors associated with closely spaced pregnancies reported as a cause of abrupt weaning in our study and in the literature [8] [13] . As for the health workers who took part in our study, they were all male, doctors (81.82%) with more than five years’ professional experience in only 9.09%, and 90.90 of them had received IMPAM and infant and young child feeding (IYCF) training. A similar trend was reported in the study conducted by these authors [8] [13] [14] , but contrary to the data reported by Sangho H et al. [15] . Since 2022, there has been a major change in the capacity-building of health workers in the management of malnutrition, as almost all providers at operational level have been trained in the revised protocol, as well as in several other approaches and normative documents on nutrition.

The majority of children (82.35%) were not prepared for weaning, although this is necessary to prevent the onset of acute malnutrition. In our context and during the ablactation period, weaning foods are generally not sufficient in quantity and quality to cover growth needs, leading to deficiencies and greater fragility of the children in the face of infections that aggravate their immune deficiencies. Another explanation is that, in most cases, ablactation is not carried out under ideal conditions. In most cases, ablactation is not carried out under ideal conditions: it is usually decided abruptly, without taking the necessary measures for diversification, or it is carried out during an illness or because of a new pregnancy. In our study, the main cause of abrupt weaning was close pregnancy. These practices make it difficult for the child to cross the threshold, thus upsetting the nutritional balance.

With regard to feeding methods, the WHO protocol for the integrated management of acute malnutrition (IMPAM) [16] recommends observing breastfeeding before giving ready-to-use therapeutic food (RUTF), but in our study 58.62% of mothers of malnourished children did the opposite. This behaviour can be explained by the lack of knowledge among mothers and caregivers about the correct use of RUTF. However, after this study, cascade training of health workers on the revised IMPAM protocol and of several women’s associations or groups on infant and young child feeding (IYCF) was carried out thanks to the support of financial and technical partners.

This evaluation revealed that anthropometric equipment, national protocols for the management of acute malnutrition, and nutritional drugs and inputs were available and in good condition. This positive finding is the fruit of the efforts made by health authorities and technical and financial partners to improve the conditions under which malnourished children are cared for in our health facilities. However, our study revealed that two health centers out of ten, or 18.18%, had no functional drinking water points. This alarming finding could be linked to poor governance and insufficient community ownership of health facilities. Despite this relative lack of governance, the facilities had task allocation notes for staff, a schedule for organizing nutrition education sessions and periodic (monthly, quarterly and annual) management and coordination meetings.

Program coverage for integrated management of acute malnutrition (IMPAM).

The number of malnourished people admitted to the program was significantly higher than expected in seven out of ten facilities. The fact that the majority of facilities exceeded the expected rate is explained by the fact that patients from outside the health areas concerned are also admitted to the program, for a variety of reasons, including the internal displacement of people due to insecurity. All this has been exacerbated by poverty and poor harvests in a context of inter-community conflict that has existed in the region since 2012.

1) Performance indicators for integrated management units for acute malnutrition ( Table 7 )

The results of our study revealed that the performance indicators (cure rate and death rate) were acceptable in the various integrated management units for acute malnutrition. However, the drop-out rates recorded in URENAM and URENAS exceeded the minimum acceptable threshold (15%) according to the National protocol for the integrated management of acute malnutrition in Mali (IMAM) [17] , as our series reported 19% and 16% drop-out rates in URENAM and URENAS respectively. These rates are significantly lower than those reported by some authors in the literature [10] [13] [18] - [20] . The dropout rates above the reference threshold observed in our study can be explained, on the one hand, by the long distance between villages and health facilities in areas that are geographically difficult to access, thus corroborating the results of the SQUEAC (Semi quantitative Evaluation of access and Coverage) survey carried out in July 2016 in Douentza, in which distance emerged as a major barrier to access to treatment and regular follow-up of children. In addition, several localities in the Douentza health district had experienced blockades by unidentified armed men, seriously impacting access to healthcare and health services in general. In the work of A.T. Coulibaly, distance was also highlighted as an important factor in the abundance of treatment, since according to the study, almost all (96%) mothers of children travelled more than 5 km to reach the nearest health service [8] . In view of these findings, strengthening communication for development (C4D) activities is one of the communication approaches needed to help reduce this rate. This communication, focused on behavioral change, is an “interactive process for the development of specific messages and approaches using various communication channels with the aim of encouraging and supporting appropriate positive behaviors”. These desired behaviors include adherence to treatment, and respect for therapeutic follow-up. At UNICEF, nutrition is one of the areas in which C4D principles and methodologies have been successfully integrated [21] . Other approaches, such as the use of community relays and community platforms, could play an important role in raising community awareness.

2) Mothers’ satisfaction with the integrated management program for acute malnutrition

Satisfaction signals the disappearance of the tension of desire, the dissipation of displeasure, of the psychological pain that generally accompanies a state of lack or need. In most cases, satisfaction can be associated with the concept of joy [22] . In this respect, the majority of mothers or guardians questioned (93.48%) were satisfied with the care provided for their children, and 67.39% thought that the integrated malnutrition management program was very good. These data are higher than those reported by Konate S et al. in their study evaluating the management of severe acute malnutrition in the pediatric ward of the Gabriel Touré University Hospital (CHU) in Bamako, Mali [10] . In our study, mothers of malnourished children were most satisfied with the welcome they received and the free care they received.