The effect of interrupted 5-day training on Integrated Management of Neonatal and Childhood Illness on the knowledge and skills of primary health care workers

Abstract

The conventional 8-day Integrated Management of Neonatal and Childhood Illness (IMNCI) training package poses several operational constraints, particularly due to its long duration. A 5-day training package was developed and administered in an interrupted mode of 3 days and 2 days duration with a break of 4 days in-between, in a district of Haryana state in northern India. Improvement in the knowledge and skills of 50 primary health care workers following the interrupted 5-day training was compared with that of 35 primary health care workers after the conventional 8-day IMNCI training package. The average score increased significantly (P < 0.05) from 46.3 to 74.6 in 8-day training and from 40.0 to 73.2 in 5-day training.

Knowledge score improved for all health conditions, like anaemia, diarrhoea, immunization, malnutrition, malaria, meningitis and possible severe bacterial infection, and for breastfeeding in 8-day as well as in 5-day training. Average skills score for respiratory problems increased from 38 to 57 in 8-day training and from 41 to 91 in 5-day training. Corresponding increases in skill scores for diarrhoea assessment were from 28 to 67 and 48 to 75, and for breastfeeding assessment from 33 to 84 and 42 to 86 in 8-day and 5-day training, respectively. Average counselling skill score also rose from 42 to 89 in 8-day and from 37 to 70 in 5-day training. A direct cost saving of US$813 for a batch of 25 trainees and an indirect cost saving of 3 days per trainee and resource person makes the interrupted 5-day IMNCI training more cost-effective.

Neonate, child, training, breastfeeding, anaemia, neonatal sepsis, malnutrition, counselling

KEY MESSAGES

• We found that 8-day and 5-day training programmes in Integrated Management of Neonatal and Childhood Illness (IMNCI) were both effective in improving the knowledge and skills of primary health care workers in a district in India.

• The full content of the conventional 8-day training can be administered in 5 days by re-scheduling the curriculum without compromising on learning activities.

• In comparison with the conventional 8-day training, the 5-day IMNCI training package is equally effective, less costly and less time-consuming.

Introduction

Integrated Management of Childhood Illness (IMCI) is a strategy for reducing mortality among children under the age of 5 years (Tulloch 1999). This strategy was developed in a stepwise fashion to address the limitations of disease-specific child health programmes (Claeson and Waldman 2000). The IMCI strategy includes the provision of antibiotics for pneumonia and dysentery, antimalarials for fever, oral rehydration therapy for prevention and treatment of dehydration due to diarrhoea and use of vitamin A for treatment of measles (Gove 1997). Nutritional counselling including promotion of exclusive breastfeeding is an integral component of this strategy.

The IMCI strategy is applied only to children from the age of 1 week to 5 years, and the generic training package is of 11 days duration, in which neonatal problems are discussed in the last two days of the training (Gove 1997). In most of the countries where IMCI has been implemented, a neonatal component was not included. Operational research in many countries has revealed several difficulties, such as finding suitable trainers for a long duration, an appropriate number of ill children for demonstration, the absence of health workers for a long period from their work stations, and large financial resources required for the training. (Lambrechts et al.1999; Kumar 2003). Thus, some countries started using a 7-day training package (Lambrechts et al.1999), though there was no formal recommendation from the World health Organization (WHO) for this change.

In 2000, the Government of India incorporated an early neonatal care component (first week of life) in the IMCI and rechristened it as Integrated Management of Neonatal and Childhood Illness (IMNCI). The 11-day training package recommended by WHO and UNICEF was adapted in India for physicians working at primary health centres. Major changes were: reduction of the training duration to 8 days, addition of an early neonatal care component, and restructuring the training programme such that at least 50% is spent on young infants (0–2 months), and reversing the order of training such that the newborn package is covered first (Chaturvedi and Chaturvedi, undated). This package was further adapted by simplifying the guidelines and adding a home visit component. In India, Auxiliary Nurse Midwives (ANMs) and Anganwadi Workers (AWWs) need to be trained to implement the IMNCI package as they work in villages and urban slums close to the people. ANMs receive 18 months pre-service training to deliver maternal and child health care services, family planning and immunization to a population of about 5000. AWWs, who serve a population of about 1000, are trained for 3 months before placement which includes supplementary nutrition, non-formal pre-school education, counselling and treatment of minor ailments.

Full coverage with 8-day IMNCI training would require a long timeframe in India even if planned trainings are conducted regularly. Availability of facilitators, participants and accommodation for 8-day IMNCI training is difficult. Experience from multi-country evaluation studies has also shown problems in operationalization of IMCI training and expansion of the programme to the country level (Bryce 2005). According to the guidelines for introducing IMCI in countries (WHO and UNICEF 1999), India is in the early implementation phase. Before expanding the IMNCI activities, it is pertinent to revisit the training strategy which is the first resource-intensive bottleneck. An alternate training strategy is required which is of lesser duration, is less costly and leads to minimal disruption of routine health care activities, but which is equally effective as the 8-day training. Therefore, we designed a less costly, interrupted 5-day IMNCI training package and tested its effectiveness in comparison with the conventional 8-day training.

Methods

This study was carried out in Panchkula district of Haryana state in northern India, which was purposively selected due to its easy access from our Institute, availability of paediatricians in the district hospital, outpatient and inpatient facilities, and occurrence of about 10 to 15 deliveries every day. This district is divided into two administrative blocks, each having one community health centre (CHC). Raipur Rani CHC is about 30 km and Kalka CHC is about 12 km from the district headquarters. There are three primary health centres (PHCs) and 19 Sub-Centres (SCs) in Raipur Rani Block, and four PHCs with 28 SCs in Kalka block.

The study population comprised all ANMs and AWWs of the selected district. All ANMs working in the sub-centres and one AWW per sub-centre, preferably from the village where the sub-centre was located, were nominated for the training by the State Health Services, Haryana according to their training plan, so as to train them as a team. The recruitment of participants to a particular training batch, i.e. 5-day or 8-day, was not controlled by the investigators.

Equivalence trial design was used for sample size calculation. It was assumed that the knowledge and skill scores would increase from 50% to 90% after the training (Abdel et al.1998). A difference of 15% in the two training types was considered as equivalence. At 95% confidence interval and 80% power, a sample size of 50 was required for each training type (Armitage et al.2002). Therefore, a training plan was prepared to train 50 trainees using the conventional 8-day course and 50 trainees by the interrupted 5-day training package in order to train all ANMs and AWWs working in the sub-centre villages of the selected district.

The conventional 8-day training package was delivered continuously in 8 days as per the national guidelines. The 5-day training package was delivered in two parts with a break of 4 days in-between. These 4 days were used to complete on-the-job training assignments as well as to carry out the health care service responsibilities, e.g. antenatal care and vaccination. The first part of the training was delivered on Thursday, Friday and Saturday, and second part was delivered on Friday and Saturday of the next week. Standard training materials developed by the Government of India in collaboration with UNICEF and WHO were used in the conventional 8-day as well as in the 5-day training. The time taken to complete different sessions in the training was recorded. Eight day training typically started in the afternoon on the first day and finished by lunch time on the last day. Lunch and tea breaks consumed about 2 hours/day. Training time was from 9 a.m. to 5 p.m. In the 5-day training, time for lunch and tea breaks was reduced. Lunch break was for 30 minutes, and tea was served during the training sessions but breaks of about 5 minutes were given to participants to freshen up. Training time was maintained from 9 a.m. to 5 p.m. However, on 2 days the training session continued up to 5.20 p.m.

Training time allotment for the course content in terms of number of module chapters, video exercise and case demonstration, chart reading, group discussion and role-play remained unaltered in the 5-day training. Informal discussions in the group were minimized. Due to selection of a nearby village/slum for field visits, travel time in the 5-day training was also less. However, the type of cases shown, allotted and discussed with participants was similar to the 8-day training. In the 5-day training, participants were sent back to their respective place of posting after the initial 3 days of training. They performed their routine health care activities at their place of posting for 4 days. In this period, they were given assignments to examine children in their area based upon IMNCI guidelines. On return to the second part of training, these assignments were discussed. Thus, loss of field time in the 5-day training was compensated with the assignment-based training strategy.

The training schedule was also changed in the 5-day training; morning time was utilized for classroom study until cases were arranged in the hospital or in the community. Module study was re-scheduled in such a way that major course content was covered in the first 3 days and many role plays were shifted to the second part of the 5-day training. Time spent on classroom sessions was 18.2 hours in 8-day and 17.2 hours in 5-day training. Similar time was spent on module reading (8 hours), video demonstration and exercises (3 hours), and role plays (3 hours) in both the trainings. An extra 1 hour was spent on group discussion in the 8-day training (4.2 hours) than in the 5-day (3.2 hours) training, and more time was spent on clinical case examination/community visits in the 8-day training (16 hours) than in the 5-day training (10 hours).

In 8-day training there were three hospital sessions and four community sessions, while in 5-day training there were two of each. In both training programmes, participants worked on cases in pairs during clinical sessions. Every participant worked on a minimum of four cases in each clinical session. All participants shared bedside discussion of all types of case identified for the training session. In 8-day training, as per the protocol, participants practised only for the clinical condition that was discussed in the classroom the previous day. For example, on day 2 they worked on possible severe bacterial infection. Breastfeeding assessment was conducted on day 4. Complete newborn assessment was done on day 4 and day 5 in a community setting. However, in the 5-day training, participants completed evaluation of a newborn including breastfeeding assessment on day 2 and they completed assessment of a child above 2 months of age on day 3. The participants in 5-day training were then asked to assess at least five children in their own village clinic or during home visits in the 4-day break. These cases were discussed on day 4 after they joined the second part of the training. Two community-based sessions were organized on day 4 and day 5.

Three IMNCI trainers—one paediatrician and two community physicians—facilitated the training sessions in each group. The same facilitators conducted the conventional 8-day training and the shorter 5-day training except for one paediatrician who could facilitate only one batch of 8-day training, but another paediatrician was available for the other training batches.

Knowledge and skills were assessed before and after the training. A multiple choice questionnaire was administered before and after the training to assess participant knowledge, which carried a maximum score of 28. It had questions to test knowledge about anaemia, breastfeeding, diarrhoea, immunization, malnutrition, malaria, meningitis and possible severe bacterial infection. Facilitators tested skills of all participants for counting respiratory rate, detecting chest in-drawing, looking for dehydration, assessment of breastfeeding technique, and counselling. Video case demonstration, role play and dummy client examination were used for skill assessment. Skill observation checklists in four domains, i.e. respiratory, dehydration, breastfeeding and counselling, were used that carried a total score of 26. At the end of training, a clinical problem-based questionnaire with a maximum score of 30 was also used to assess skill in consulting the modules and treatment charts for making decisions about various disease conditions.

The same set of questionnaires and checklists were used in the pre- and post-test assessment. Participants were instructed not to put any identifying information on the questionnaire. They were assured that confidentiality would be maintained and individual scores would not be revealed to the health authorities.

Data analysis was done using Epi Info 2000 computer software. Since the maximum scores for knowledge (28), skill assessment by clinical problem-based questionnaire (30) and skill evaluation through observation checklist (26) were different, the score of each participant for each of the knowledge and skill assessment methods was converted to a relative scale with a maximum score of 100. For example, if the score of a participant was 10 out of a maximum score of 28, then the score of the participant on a common scale having a maximum score of 100 would be 35.7 [(10/28) × 100]. An unpaired t test was used to find any statistically significant rise in values after the training. Since individual identifiers of PHC workers were not recorded on evaluation sheets, it was not possible to track the scores of individual's in order to perform a paired t test.

The cost of each 8-day and 5-day training course was computed by adding up expenditure incurred on travel and daily allowances by participants as well as facilitators, contingency money per participant to meet the cost of stationery, refreshments, food, transportation costs for community visits, and costs of hiring audio-visual equipment. Expenses on the use of buildings and other office equipment and furniture, electricity etc. were not computed as these will be available free of cost in public institutions. The cost of running one training batch for 25 PHC workers for 8-day as well as interrupted 5-day training was calculated (see Table 3).

Table 3

Estimated expenses for a course of 8-day and interrupted 5-day IMNCI training

Expense category	8-day training		5-day training
	US$	Indian Rupees	US$	Indian Rupees
Travelling allowance
External facilitators @ Rs 4000/trip * 2 persons	200	8000	400	16 000
Local facilitators @ Rs 200/trip/day * 4 persons	160	6400	100	4000
Participants @ Rs 200/trip * 25 persons	125	5000	250	10 000
Daily allowance
Participants @ Rs 125 per day * 25 persons	625	25 000	391	15 625
Facilitators @ Rs 500 per day * 6 persons	600	24 000	375	15 000
Petrol/oil for field visit @ Rs 2000 per visit	200	8000	100	4000
Accommodation for external facilitators @ Rs 1000/day * 2 persons	400	16 000	250	10 000
Audio-visual aid rental @ Rs 1000 per day	200	8000	125	5000
Contingency @ Rs 100 per person per day * 25 persons	500	20 000	313	12 500
Subtotal	3010	120 400	2304	92 125
Administrative charges @ 15% of subtotal	451	18 060	345	13 819
Total	3461	138 460	2649	105 944
Cost per participant	138.0	5538	106	4238

Results

Thirty-five participants attended the 8-day training and 50 participants joined the 5-day training. Each type of training was completed in two batches. In the 8-day training there were 23 (10 ANMs and 13 AWWs) and 12 participants (11 ANMs and 1 AWW) in the first and second batch, respectively. Attendance in the first and second batch of the 5-day training course was 26 (13 ANMs and 13 AWWs) and 24 (12 ANMs and 12 AWWs), respectively. In one of the 8-day training batches, participation of AWWs was less because their invitation was not timely.

The socio-demographic characteristics of IMNCI training participants who attended the 5-day training were statistically similar to those who attended the 8-day training. Mean age of the participants was 38.8 years in 8-day and 39.3 years in 5-day training. Average service experience was 15.9 and 13.9 years in 8-day and 5-day training, respectively. About 65.8% of the participants were educated up to 10th standard in 8-day training compared with 62.0% in the 5-day training (P > 0.05).

Before IMNCI training, the average knowledge and skill scores of the participants who attended the 8-day and 5-day training were similar (P > 0.05). After the training, average score increased significantly (P < 0.001) from 46.3 to 74.6 in 8-day training and from 40.0 to 73.2 in 5-day training. The rise in average score was statistically similar in both training types (P > 0.05). Average knowledge score rose significantly (P < 0.001) from 52.5 to 73.2 in 8-day training and from 51.7 to 72.1 in 5-day training, whereas average skill score rose significantly (P < 0.001) from 36.7 to 90.9 in 8-day training and from 28.5 to 83.3 in 5-day training (Table 1). Further explorative analysis revealed that the rise in average skill score was substantial in both the 8-day and the 5-day training for both categories of PHC workers (Table 1), and rise in average knowledge score was also significant for each type of knowledge and skill domain tested (Table 2).

Table 1

Pre- and post-training knowledge and skill scores in 8-day and interrupted 5-day IMNCI training

Participants	Training type	N	Pre-training		Post-training
			Mean	95% CI	Mean	95% CI
Knowledge and skills
All	8-day	35	46.3	43.6–49.0	74.6*	72.3–77.0
	5-day	50	40.0	36.8–43.3	73.1*	71.5–74.8
Auxillary Nurse Midwife	8-day	21	39.6	27.3–51.8	90.4*	84.4–96.3
	5-day	25	28.3	20.1–36.4	82.5*	73.8–91.1
Anganwadi Worker	8-day	14	28.8	18.7–38.8	84.4*	76.0–92.7
	5-day	25	52.5	49.5–55.4	72.3*	68.4–76.1
Knowledge
All	8-day	35	27.4	23.8–31.0	81.6*	79.0–84.1
	5-day	50	46.3	43.6–49.0	74.6*	72.3–77.0
Skills
All	8-day	35	40.0	36.8–43.3	73.1*	71.5–74.8
	5-day	50	39.6	27.3–51.8	90.4*	84.4–96.3

*P < 0.001.

Table 2

Knowledge and skill score for various health conditions in 8-day and interrupted 5-day IMNCI training

Health conditions	Training type	Pre-training		Post-training
		Mean	95% CI	Mean	95% CI
Knowledge assessment
Anaemia	8-day	54.2	47.6–60.7	91.4**	83.2–99.5
	5-day	72.0	66.3–77.6	86.0*	77.8–94.1
Breastfeeding	8-day	47.9	45.1–50.6	63.5**	59.7–67.2
	5-day	59.1	54.4–63.7	61.6**	55.3–67.8
Diarrhoea	8-day	50.8	44.8–56.7	73.1**	66.8–79.3
	5-day	54.8	48.3–61.2	70.8**	61.0–80.5
Immunization	8-day	79.0	68.1–89.8	86.0	76.2–95.7
	5-day	52.5	42.6–62.3	83.7**	74.7–92.7
Malnutrition	8-day	55.9	48.0–63.7	84.0**	77.4–90.5
	5-day	57.0	50.0–63.9	79.5**	72.7–86.2
Malaria	8-day	28.5	25.2–31.7	80.0*	66.9–93.0
	5-day	20.0	16.7–23.2	80.0**	69.1–90.8
Meningitis	8-day	35.9	27.7–44.0	70.0**	60.8–79.1
	5-day	47.5	37.7–57.2	64.5**	58.0–70.9
Possible severe bacterial infection	8-day	56.0	50.1–61.8	64.4**	59.6–69.1
	5-day	43.3	39.3–47.2	60.8**	54.9–66.6
Skills assessment
Respiratory	8-day	44.2	34.8–53.5	90.6**	85.5–95.6
	5-day	39.0	30.6–47.3	87.4**	81.1–93.6
Diarrhoea	8-day	33.9	24.8–42.9	95.2**	91.3–99.0
	5-day	31.3	24.1–38.4	75.9**	69.0–82.7
Breastfeeding	8-day	19.4	16.1–22.6	94.8**	90.8–98.7
	5-day	15.2	12.2–18.2	90.4**	86.4–94.3
Counselling	8-day	43.2	37.8–48.5	83.2**	77.9–88.4
	5-day	28.7	23.2–34.1	79.7**	74.0–85.3

**P < 0.001, *P < 0.05.

Total expenditure for conducting 8-day training for a batch of 25 participants was estimated to be US$3462, whereas for conducting 5-day training it was estimated as US$2649. Thus there is saving of US$813 per training batch or US$33 per participant (Table 3).

Discussion

IMCI strategy has been implemented in more than 100 countries. However, only a few studies have documented the impact of the IMCI training programme on knowledge and skills of health workers. The 11-day IMCI training package proposed by WHO in the 1990s had no direct scientific evidence of its adequacy for all countries and regions, but at that time there was a consensus that this was the best model to be proposed to countries to assure that trainees would perform well and retain the skills obtained from the training. For various operational reasons, although training duration was reduced to 8 days in many countries, there was no direct evidence of its impact on the knowledge and skills of the participants. Our study provides evidence for the efficacy of both an 8-day and a 5-day training package in improving knowledge and skills of PHC workers, i.e. ANMs and AWWs, in a district of India.

In our study, overall average knowledge scores in 8-day as well as in 5-day training rose to above the 70% level. In another study in rural Haryana (Anand et al.2004), the knowledge of workers on diseases and their management improved after 4-day training but reached a plateau at a 50% score. Joshi (2006) demonstrated a significant 23 point rise in knowledge score following a 3-day training course on maternal and child health among urban health community workers. Abdel et al. (1998) showed a 40% rise in knowledge of health workers in Saudi Arabia following continuing medical education regarding childhood illness.

In our study, average skill scores increased by about 54% in both training types. In Uganda, health workers’ skills regarding correct assessment, classification and treatment and counselling of children improved by 20% after the 8-day training package compared with a control area without training (Pariyo et al.2005). Marked improvement also occurred among health workers after IMCI training in South Africa (Chopra et al.2005). In West Africa, health workers’ skills to assess sick children improved up to 85% following IMCI training (P < 0.05) (Kelly et al.2001). Thus, the rises in average knowledge and skills in our 8-day and 5-day training packages were comparable with other studies elsewhere.

The method of evaluation in the 5-day and the 8-day training was the same. The pre- and post-test questionnaires used in the study were also the same. The study hypothesis was also not known to the participants. No additional motivation was given to the participants in 5-day training to prove themselves and resort to home studies or other means to gain the knowledge. Participants with lower baseline scores may have gained more compared with others and achieved comparable endline scores. However, in order to maintain confidentiality, questionnaires were completed anonymously and thus it was not possible to track individual performance. Differential participant mix in the two training programmes may bias the study. In one of the 8-day training batches, the number of AWWs was less. However, mean scores at baseline and endline were found to be similar in both groups. Average scores of ANMs and AWWs in both training programmes at pre- and post-test also showed no statistically significant difference. Training by different trainers in the two trainings can also introduce bias. However, in our study all trainers were the same except for one paediatrician who was available for only one 8-day training course, and subsequently felt unable to devote the time required for the training. Another paediatrician remained available throughout the course. Achievement of equivalent scores in 5-day training without a paediatrician facilitating (though available) further makes 5-day training more economical and practical as the opportunity cost of a paediatric trainer from a busy tertiary care hospital is usually high.

Studies have documented that the performance of health service providers with respect to the diagnosis and treatment of specific childhood diseases does improve after IMCI training (Amaral 2004; Amorim et al.2008). Training also helps in managing children at lower cost (Amorim et al.2008). However, training alone is not sufficient to reduce infant mortality rates. The other two components of IMCI, i.e. community involvement and health system development, need to be addressed simultaneously (Amaral 2005). Moreover, as retention of knowledge and skills deteriorates over time, refresher training is required (Chaudhary 2005). Retention of knowledge and skills may also be different for 8-day and the new 5-day training, therefore follow-up studies should be undertaken to document any difference between participants in these two training programmes.

The 5-day IMNCI training package was more economical compared with the 8-day package. Considering that there are at least 179 000 ANMs and Lady Health Visitors posted in existing sub-health centres, primary health centres and community health centres (Government of India 2007), and that 40 000 AWWs are working in the existing Anganwadi centres (ICDS, undated), 8760 training sessions would be required to train these PHC workers in India. A direct saving of at least US$813 per training course would mean a direct saving of US$7.13 million, and an indirect saving of 657 000 person-days for participants and 157 680 person-days for facilitators at the country level. Savings with 5-day training would increase further if rental costs for using training premises, furniture, and electricity etc. are also considered. Further, the design of 5-day training allowed a break of 4 days between the two parts of the training to continue routine health care delivery of very important health care services such as immunization days (Wednesdays) and antenatal days (Tuesdays).

Organizing the interrupted 5-day training may incur some operational difficulties. The training venue has to be prepared twice, and there are chances of dropout of participants as well as facilitators. A well-motivated training coordinator is required to keep the participants engaged, managing the tasks within the allotted time and allowing extension of learning time beyond the scheduled training hours, if needed. However, despite these operational difficulties, interrupted 5-day training was not considered a burden by either the participants or the facilitators.

We can conclude that it is feasible to administer the entire learning contents of the conventional 8-day training in 5 days by re-scheduling the curriculum without compromising on learning activities. In comparison with the conventional 8-day training, the 5-day IMNCI training package is equally effective, less costly and less time-consuming. The proposed 5-day IMNCI training strategy should be applied on a wider scale only after testing its feasibility in diverse regional, cultural and logistical environments as health systems differ considerably both within India and in other developing nations.