Menu
Menu
Menu
Menu
Background: Severe acute malnutrition (SAM) is a major determinant of childhood mortality and morbidity. Although integrated community case management (iCCM) of childhood illnesses is a strategy for increasing access to life-saving treatment, malnutrition is not properly addressed in the guidelines. This study aimed to determine whether non-clinical Community Health Workers (called Community-Oriented Resource Persons, CORPs) implementing iCCM could use simplified tools to treat uncomplicated SAM. Methods: The study used a sequential multi-method design and was conducted between July 2017 and May 2018. Sixty CORPs already providing iCCM services were trained and deployed in their communities with the target of enrolling 290 SAM cases. Competency of CORPs to treat and the treatment outcomes of enrolled children were documented. SAM cases with MUAC of 9 cm to < 11.5 cm without medical complications were treated for up to 12 weeks. Full recovery was at MUAC≥12.5 cm for two consecutive weeks. Supervision and quantitative data capturing were done weekly while qualitative data were collected after the intervention. Results: CORPs scored 93.1% on first assessment and increment of 0.11 (95% CI, 0.05–0.18) points per additional supervision conducted. The cure rate from SAM to full recovery, excluding referrals from the denominator in line with the standard for reporting SAM recovery rates, was 73.5% and the median length of treatment was 7 weeks. SAM cases enrolled at 9 cm to < 10.25 cm MUAC had 31% less likelihood of recovery compared to those enrolled at 10.25 cm to < 11.5 cm. CORPs were not burdened by the integration of SAM into iCCM and felt motivated by children’s recovery. Operational challenges like bad terrains for supervision, supply chain management and referrals were reported by supervisors, while Government funding was identified as key for sustainability. Conclusion: The study demonstrated that with training and supportive supervision, CORPs in Nigeria can treat SAM among under-fives, and refer complicated cases using simplified protocols as part of an iCCM programme. This approach seemed acceptable to all stakeholders, however, the effect of the extra workload of integrating SAM into iCCM on the quality of care provided by the CORPs should be assessed further.
Using a sequential multi-method design [13], the capacity of non-clinical CORPs to use a simplified protocol for the treatment of SAM at the community level and the treatment outcomes of children who received treatment through this program were assessed and the findings complemented with opinions of key stakeholders on the acceptability of the treatment approach. The study was implemented in Mariga and Rijau, two of the six local government areas (LGAs) in Niger state where iCCM was being implemented at the time with a projected total under-five population of 104,912 [14]. Most families in these communities are poor, illiterate and lack access to basic social amenities and health care. The selection of the LGAs was informed by an unpublished report of a mass screening activity for malnutrition carried out in the state in 2016 which found prevalence of SAM to be 10% in the project sites [15]. The simplified SAM treatment protocol and tools used in South Sudan in 2016 [16] were pretested by officials of Niger State Ministry of Health, Malaria Consortium and IRC project staff with selected CORPs in Niger State in July 2017 and adapted based on their feedback, particularly the mid-upper arm circumference (MUAC) tape, weighing scale, calculator and patient register. Thus, a series of revisions were effected to make the tools more user-friendly, compliant with the cultural context of the study areas and Nigerian SAM treatment protocol. The MUAC tape is a colour-coded malnutrition screening tool that measures the circumference of the left upper arm at the mid-point between the tips of the shoulder and elbow of a child. A child with MUAC on the green zone is well-nourished while yellow zone indicates moderate acute malnutrition (MAM) and the red zone indicates severe acute malnutrition (SAM). The red (SAM) zone on the simplified MUAC used for the South Sudan study was further divided into three (pink, bright red and deep red) in order of severity. The weighing scale determines the weight and corresponding daily quantity of RUTF needed for treatment. The weighing scale is calibrated into red (referral) and other zones with dots corresponding to number of RUTF sachets needed for daily treatment. The calculator helps to determine the weekly number of RUTF and the register is used for documenting bio-data and treatment date for the patient. Key revisions on the tools included an amendment of the colours of the MUAC tape and weighing scale by changing referral colours to ‘bright red’ to align with iCCM colour codes for danger signs and referrals; alignment of the treatment duration with the national community-based management of acute malnutrition (CMAM) guidelines that has maximum treatment period of 12 weeks instead of 16 weeks contained in the South Sudan protocol; and use of context-specific images in the simplified protocol. (Kindly refer to the supplementary files for more details on algorithm, tools and graphics.) A minimum of 290 children were expected to be treated by 60 CORPs over a period of 6 months based on an estimation using the Sphere Humanitarian Standard (SHS) of 75% recovery rate for SAM. A one-sample non-inferiority test was carried against this rate with the assumption that a rate that is not more than an absolute 10% lower than the SHS would be considered non-inferior, with an alpha value of 0.05, a power of 0.90 and a loss to-follow-up rate of 10%. The latter was defined as loss to follow-up caused by the supply side (e.g. CORPs dropping out of the programme or stock-outs of ready-to-use therapeutic food (RUTF)). CORPs already trained on iCCM and providing treatment services in their homes for malaria, diarrhoea and pneumonia as volunteers in hard-to-reach communities for a minimum of 2 years, were selected and trained on the simplified protocol. Each CORP had between 115 and 200 under-five children within the catchment area. The CORPs were trained in batches over 6 days with a trainer to trainee ratio of 1:3 and strong focus on the use of the simplified SAM protocol and its integration with the iCCM treatment algorithm. CORPs’ subsequent participation in the pilot was determined by a post-training assessment score of 80 % and more. Weekly supervision was provided by community health extension workers (CHEWs) who were based at the primary healthcare facilities and already iCCM supervisors. The supervisors were also trained and in turn supervised by the LGAs’ health teams. A total of 400 cartons of RUTF, additional doses of amoxicillin and albendazole were procured and distributed to CORPs for treating SAM cases. Commodity logistics was built on the existing iCCM commodity supply mechanism, however, to store the RUTF securely, metal boxes were provided to CORPs. Supervisors were responsible for monitoring stock levels and restocking the CORPs. Prior to commencement of the intervention, the demand creation activities of iCCM were strengthened to sensitise members of communities on the additional nutrition services available with the CORPs. Informed consent was received from caregivers accessing iCCM services and whose children were malnourished. Admission to CORP’s nutrition treatment was based on a modified colour-coded MUAC strip. Children were assessed using the iCCM algorithm to either identify danger signs and refer, or treat based on the protocol. Any child screened and found not to have any iCCM danger signs but who fell in the severe ‘malnutrition zone’ of the MUAC (red or pink), was given an appetite test to further screen for enrolment eligibility. If the child passed the appetite test by being able to eat as much as one quarter of the RUTF pack, s/he was enrolled in the study. A failed appetite test resulted in referral to the appropriate health facilities for treatment. Upon enrolment, CORPs administered amoxicillin and albendazole to each SAM case according to the simplified protocol. To determine the RUTF doses required per day, the child was weighed with the Salter scale overlaid with a dosage chart that guided the CORP on the number of sachets of RUFT to administer, depending on where the weight indicator fell on the chart. Rather than reading weight as figures in kilograms, the scale indicated the number of sachets of RUTF needed per day. The seven-day dosage was then calculated with the aid of the simplified calculator, a rectangular strip with 7 pockets, one for each day dose. Using a flipchart, CORPs counselled caregivers on how to administer the RUTF and other medications at home, adhere to the daily dosage, maintain good hygiene and return the following week to continue treatment, except if the condition of child got worse before the next appointment. Each SAM patient encounter was recorded in a register and CORPs followed up defaulting enrolees with home visits, recorded children’s progress every week and discharged as appropriate based on possible outcomes listed in Table 1. The maximum treatment period for any admitted case was 12 weeks. Admission and discharge criteria and associated treatment actions CORPs’ demographic information was collected at the outset of the project by MC Research Officers, while children’s demographic information was collected by the CORPs on patient register and by supervisors on supervision checklist used for assessing compliance of CHWs with the simplified treatment protocol. Thereafter, progress data for enrolees (i.e. malnutrition status, number of weeks in treatment and treatment outcomes) and stock levels of commodities were collected using treatment registers and stock management tools respectively. Each CORP was supervised and assessed weekly by their supervisors on ability to use the simplified tools and compliance with the treatment protocol. There was an average of three CORPs to one supervisor. Data analysis focused on treatment outcomes (% recovered, % defaulted, % non-response, % death), and number of weeks in treatment and data was stratified by child background characteristics including age and severity of malnutrition at enrolment. Single and multiple regression models were run to determine associations between child/CORP characteristics and treatment outcomes (recovered vs. not recovered as reference group) using a modified Poisson approach to arrive at an appropriate risk ratio estimate, which is more interpretable than odds ratio [17]. Stata version 14 was used for analysis. The qualitative component of the study sought to probe for in-depth information from the stakeholders on the acceptability of the intervention. A total of eight Focus Group Discussions (FGDs) were conducted. There were four FGDs conducted with CORPs, two with supervisors and two with caregivers. Study communities were clustered into four groups based on contiguity and participants for each FGD were randomly selected from each cluster. Similarly, a total number of 17 in-depth interviews (IDIs) were conducted with CORPs; caregivers; policy makers and programme implementers. The respondents for the IDI were CORPs highest and lowest performers based on assessment scores while the IDI respondents among caregivers were those with cured, defaulted, non-response and referred treatment outcomes. The policy makers for IDI were from the National, State and LGA levels while programme managers were from both Government and non-Government agencies. As Table 2 shows, roughly equal numbers of CORPs were assessed in the two LGAs, the vast majority of whom were male and aged 18–35 years. A large proportion reported they had senior secondary level education (not verified), could read without any difficulty and had been working as CORPs for between three and 4 years. Background and demographic characteristics of community-oriented resource persons There were 528 supervisory visits and performance assessments on the 60 CORPs who treated SAM cases. All (100%) CORPs with current SAM cases were supervised weekly as planned. The mean number of assessment carried out on each CORP was 10; median 10, range 1–28, IQR was 5–15. CORP performance was analysed from a maximum of twelve consecutive competency-based assessment scores. Tables 3, ,44 and and55 show the performance of the CORPs. First competency-based assessment was high at 93.1% with an increase of 0.11 (95% CI: 0.05. 0.18) points in score per additional supervision conducted. When CORPs’ performance on individual tasks in the treatment procedure was examined (Table 4), all CORPs scored more than 90 % in all tasks except prescription and recording of albendazole where the score was 82.2% (95% Ci: 71.4, 89.5). All the CORPs did excellently well in terms of giving adequate messages to caregiver on RUTF dosage Table 5). Performance score of CORP on biodata and screening for danger signs, accounting for clustering at CORP level Performance score of CORP on malnutrition treatment procedures, accounting for clustering at CORP level Performance score of CORP on malnutrition treatment messages to caregivers, accounting for clustering at CORP level Of the 303 children enrolled, complete records with all data elements were available for 288 children (95%). Information on child’s gender, mother’s age and level of education as well as other nutrition-related characteristics of the household are contained in Table 6. Background and demographic information of study enrolees Of the 303 children enrolled for treatment of SAM, treatment outcomes were available for 288 (96%). As Table 7 indicates, there was a cure rate of 73.5%, excluding referrals from the denominator in line with the standard method of calculating recovery rates in nutrition interventions. The median number of weeks in treatment for the cured children was 6.7. The most common reason for defaulting was that the caregiver did not wish to continue with the treatment or had sought alternative care elsewhere (36%). Forty-three (14.8%) were referred with over half (56%) of these cases being due to failed appetite test. Treatment outcomes Median: 6 Mean: 6.7 IQR: 5–8 Range: 4–12 Median: 6 Mean: 6.8 IQR: 5–8 Range: 4–8 Median: 3 Mean: 4.1 IQR: 2–6 Range: 2–10 aStandard definition of recovery rates in nutrition programmes Table 8 shows the adjusted risk ratios of factors affecting recovery. Children who started in the dark red MUAC zone had 31% (95% CI: 7–49%) less likelihood of recovery compared to those that started in the pink zone. Children seen by CORPs with catchment populations of 50 and above, up to less than 200, tend to have a lower chance of recovery compared to those with catchment population less than fifty. Maternal religion also found to impact recovery, which may be a proxy for socioeconomic or cultural drivers of recovery. Factors affecting treatment outcomes The FGD and IDI responses were categorised according to the following pre-determined themes: perception of malnutrition, general impression of the project, community-CORP relationship with regard to the treatment, workload of CORPs, supervision, supply chain, tools, referral mechanism and sustainability. Details of the qualitative data are contained in the supplementary file. Several caregiver FGD respondents reported that malnutrition is defined by a loss of weight or a “shrunken” look on a child, with a few mentioning that fever, by definition, is a part of malnutrition and a few others reporting that it is due to spiritual causes. It appeared from the responses that there was no clear distinction of malnutrition as a specific disease; the respondents generally described it as a state of weakness and illness. The two individuals who noted spiritual causes indicated that they realized that the illness was due to a different cause when their children eventually recovered after treatment. One IDI respondent blamed spiritual forces for her child’s failure to recover. Several respondents noted that the illness was a result of the mother getting pregnant while still breastfeeding, and others reported that, prior to receiving treatment, their families thought the illness was not curable and did not believe the treatment would work. All respondents expressed positive disposition toward the program indicating child’s recovery, free care, and shorter distance to access care as some positives. One supervisor mentioned that: “formerly, if there is a case of red MUAC, it is abandoned because there is no access to treatment.” A few caregivers reported that their family members were relieved upon hearing that there were no costs associated. One reported that her husband thought that treatment was expensive but upon notification that the treatment was free, he was very happy. Another reported that the respondent’s grandmother was already considering what personal belongings to sell in order to afford the treatment. In FGDs, CORP respondents reported being more respected by the community and that they are better recognized by the community members. A few of the CORPs also reported receiving gifts such as a chicken, grains, and money to fuel their motorcycles. Despite this expression of gratitude for the CORP treatment program, a frequently reported perception of the community was that the CORPs were biased or displayed favouritism in deciding who should or should not receive treatment. This issue was raised in both CORPs and caregiver groups where situations in which two children would be screened at the same time, one would be deemed eligible and the other not were described. This led to dissatisfied caregivers accusing CORPs of favouritism. A CORP mentioned that a caregiver felt it was her child’s right as a community member to be given RUTF while another noted that some community members were not aware that the service was meant only for malnourished children (Appendix 3 box 4). Supervisors and community health program managers all raised concerns about workload, and several of these respondents suggested the need to pay CORPs for the additional work. However, the CORPs themselves in their responses appeared to be happy with the addition of the nutrition treatment program to their tasks as this resulted in them gaining more respect from community members. Rather than requesting payment as a compensation for the extra workload, the CORPs expressed satisfaction and felt compensated by the knowledge and skill gained by treating malnourished children. During the intervention, CORP were requested to identify a convenient clinic day and time that worked with the schedules of their primary occupation (e.g. identify 1–2 days a week, limit to evening times). The CORPs stated that they did not feel burdened by the SAM treatment schedules because they were given the opportunity to choose convenient time for the treatment. The supervisors reported issues of caregivers not making timely visits or CORPs not being available at time of visiting, leaving situations in which either party is waiting for the other. Mobility for supervisors was also difficult due to long distances or bad roads. Frequency of supervision was brought up as a difficulty by program managers, given that for the iCCM program, the CORPs were only receiving quarterly supervision and the distribution of iCCM medication only happens every 2 months. Suggestions for improving supervision did not have an overarching theme, with suggestions ranging from making sure CORPs have all the tools, more financing, more manpower for supervision, having a calendar to ensure supervision visits are not missed, and reduced frequency of supervision (to monthly, quarterly). In FGDs, CORPs reported that having supervisors correct their mistakes in person was much appreciated. One respondent however indicated that the supervisors should not correct them in front of the caregivers, as it makes them look as if they do not know what they are doing (Appendix 3 box 6). CORP supervisors appeared to have some difficulty in moving the RUTF around to hard-to-reach places, as by definition iCCM is prioritised for areas which are hard to reach. Programme implementers during IDI, indicated that initial procurement of RUTF was also very challenging, given that there was no existing community level nutrition programme in the area. Despite these difficulties, only one CORP FGD had a few respondents report stock-outs, while others indicated that they never had any issues with stock. One respondent indicated that the CORP supervisor would even borrow excess stock from others to ensure there was no stock-out. Another issue raised regarding keeping stock at home was ants being attracted by empty RUTF sachets. The difficulty of calibrating the weighing scale was mentioned several times and doing MUAC measurement on an active child was also raised as an issue. A CORP suggested that a specific stand be designed for a weight scale rather than hanging it from a tree. A few people raised issues about how best to record data for individuals who do not regularly come, or in other words, how best to record a missed visit. Other suggestions included having an identification card for SAM children like they do for iCCM and uniforms for CORPs providing SAM treatment. Referrals appeared to have been difficult, given that the care provided by the CORP was free and the referral would cost money. The referral site was approximately 80 km away from the intervention sites. Specifically, with appetite test failure, one CORP mentioned that a caregiver insisted that her child be enrolled to receive RUTF, despite the explanation that the child needed to be referred. Some also appeared to perceive this as an issue of favouritism; that certain CORPs did not want to treat them, hence they referred them to the health facility. Reasons brought up by CORPs supervisors for incomplete referrals included financial limitations, ignorance, families being convinced that the child is going to die anyway if CORP were not able to cure, and belief that the cause was spiritual. Several respondents stressed the need for government ownership, and the need for results to be shared with local and village level leadership to receive buy-in and inspire collective responsibility. A few additionally suggested that the State Ministry of Health should identify a budget to invest in this program. In contrast, one respondent said that the program responsibility should remain with Malaria Consortium, as the local government does not have the capacity to sustain the program.
