Community-based interventions for diarrhoeal diseases and acute respiratory infections in Nepal

Abstract

Problem

Acute diarrhoeal diseases and acute respiratory infections (ARIs) are the most common causes of child mortality worldwide. Safe, effective and inexpensive solutions are available for prevention and control, but they do not reach needy communities.

Approach

Interventions based on research were designed to train and engage community health volunteers (CHVs) to implement a community-based control programme in Nepal. With the advent of the Integrated Management of Childhood Illnesses (IMCI) strategy, this programme subsequently emerged as a community-based IMCI but retained its mainstream activities. We reviewed and analysed policy decisions and programme development, implementation and expansion.

Local setting

Severe resource constraints and difficult terrain limit access to health-care facilities in many parts of Nepal.

Relevant changes

In districts with interventions, more cases of acute diarrhoea and of ARIs (including pneumonia) were reported. The proportion of diarrhoea cases with dehydration and the proportion of ARI cases with pneumonia were significantly lower in districts with interventions. Case fatality rates due to acute diarrhoea and the proportion of severe pneumonia among ARI cases across the country showed a significant trend towards a decrease from 2004 to 2007. Nepal has succeeded in training many CHVs and is on course to meet the Millennium Development Goal for child mortality.

Lessons learnt

The burden of acute diarrhoea and ARIs can be reduced by training and engaging CHVs to implement community-based case management and prevention strategies. Monitoring, supervision and logistical support are essential. Policy decisions based on evidence from national research contributed to the success of the programme.

Introduction

Acute diarrhoeal diseases and acute respiratory infections (ARI), especially pneumonia, remain the leading causes of child mortality globally. They are also the leading causes of disability-adjusted life years across all age groups.¹ The South-East Asia Region (http://www.who.int/about/regions/searo) of the World Health Organization (WHO) carries 30% of the global burden, and these diseases affect mostly poor and marginalized groups. Simple, safe and inexpensive interventions with proven effectiveness are now available, but low coverage has resulted in poor control of these diseases.² This article describes a strategy that involves community health volunteers (CHVs), and that has worked in Nepal. As a result of the interventions we report here, Nepal is headed towards achieving Millennium Development Goal 4: to reduce child mortality. Our findings are relevant for countries with similar demographics and high burdens of these diseases.

The context

Nepal recorded high under-five mortality averaging about 170 annual deaths per 1000 in the early 1980s, and now reports 61 per 1000.^3,4 WHO-supported programmes for the control of diarrhoeal diseases and respiratory infections started during the 1980s and reduced child mortality. More recently, Nepal developed innovations including a community-based programme for the control of ARI and diarrhoeal diseases. As a result, Nepal is now on course to meet Millennium Development Goal 4 despite many constraints and challenges. However, many other countries in the region that report higher disease burdens^5,6 may benefit from community-based interventions similar to those used in Nepal.

Approach

We reviewed the evolution of relevant health policies and programmes used during the preceding 20 years, examined the processes and events that influenced these policies and evaluated the impact of community-based interventions. Our analysis was based on interviews with past and present policy-makers and programme managers and a review of the literature, annual reports and work plans of the Ministry of Health.⁷ To evaluate impact we analysed quantitative data from the Health Management Information System. Incidences of diarrhoea and ARI in Nepal’s 75 districts were calculated with each district’s under-five population according to the 2001 national census as the denominator.

Programme development

A community-based national programme to control diarrhoeal diseases was launched in 1982. A working group oversaw planning, budgeting, management and logistics. Strategies included promoting home-based oral rehydration therapy and case management in health facilities. In 1991, district-level planning and community mobilization were included, and strategies were expanded to include oral rehydration therapy, together with oral rehydration corners in health centres and hospitals, and case management by peripheral health workers. In addition, district health managers were involved in social mobilization, programme planning and training. Diarrhoea case management was also assisted by traditional local practitioners and drug sellers.

Research based on national data established that CHVs could be trained to correctly diagnose and categorize the severity of pneumonia, as well as to improve case management at the community level.^8,9 Close monitoring, supportive supervision and proper management of logistics and supplies ensured quality of care. A working group for ARI created in 1992–1993 expanded case management of childhood pneumonia by involving women from the community who worked as volunteers. A programme to improve community-based care began with a case-control study in 1995: in two districts women volunteers participated in diagnosis and treatment, and in two other districts they participated in diagnosis and referral of cases to formal health facilities. In 1997 an evaluation supported by WHO found that the proportion of expected pneumonia cases that received correct treatment nearly doubled in the treatment group but there was no change in the referral group.¹⁰

Subsequent integration of programmes led to the establishment of a community-based ARI and diarrhoea control programme. Women – some with basic school education and some semi-literate or illiterate – are trained as CHVs by the Ministry of Health to diagnose, assess disease severity and danger signs, treat children and refer them to health facilities. This programme also provides nutritional and immunization services. Training curricula and modules based on WHO’s Integrated Management of Childhood Illness (IMCI) strategy were simplified and adapted to make them interactive and suited to participation by these groups.⁴ The value of modifying IMCI training is being recognized by many, including WHO.¹¹

In its first year the programme was launched in five districts, including two very remote districts with difficult terrain and access. The choice of sites for initial intervention was based on consensus between the Ministry of Health and Population, its development partners and district administrations. This approach ensured logistical support, monitoring and supervision, in contrast with IMCI programmes elsewhere, which often select facilities that can be easily accessed and equipped.^12,13 In 1999, this community-based control programme merged with the newly advocated IMCI strategy and acquired a new name: “community-based IMCI”. The core components and functions of the original programme, however, remained unchanged.

The programme expanded gradually to cover 48 districts by the end of 2007, with plans to cover the remaining 27 districts by the end of 2009. In districts without the programme, current facility-based care and immunization programmes continue. Similarities in the terrain and social conditions in the country’s 75 districts lead us to assume that pre-intervention patterns of disease epidemiology and risk factors are similar in districts that received and did not receive interventions. This assumption, however, has not been formally tested.

Results

The impacts of community-based interventions were evaluated on the basis of data for 2006 and 2007, before the programme was expanded to 15 more districts. The rate of reported episodes of diarrhoea was higher in the 33 districts with community-based interventions. The 377 770 diarrhoea episodes reported in a total under-five population of 1 798 668 in districts with interventions represented 0.21 episodes per child per year. In the 42 districts without interventions, the 303 049 episodes reported in a total under-five population of 1 873 982 represented 0.16 episodes per child per year (χ² for the difference in proportions = 14 186; P < 0.01).

In districts with interventions the proportion of diarrhoea episodes with some dehydration (110 956/377 770, 29.4%) was significantly lower than in districts without interventions (107 238/303 049, 35.4%; χ² = 2820.4; P < 0.01). Likewise, the proportion of diarrhoea episodes with severe dehydration was lower in districts that received interventions (3108/377 770, 0.8%) than in those without interventions (4465/303 049, 1.5%; χ² = 646.5; P < 0.01). Between 2004 and 2007 more districts were included in the programme. During this period the proportions of diarrhoeal episodes with some dehydration or severe dehydration nationwide, as well as the national case fatality rates for acute diarrhoea, showed a significant trend towards a decrease (Table 1).

Table 1. Trends in acute respiratory infection, pneumonia, diarrhoea and death from diarrhoea in Nepal, 2004–2007.

Year	Population aged < 5 years	New ARI cases			New pneumonia cases				New severe pneumonia cases				Episodes of diarrhoea			Cases of diarrhoea with some dehydration				Cases of diarrhoea with severe dehydration				Diarrhoea deaths
		No.	ECYa		No.	Percentage of ARI cases	ECYa		No.	Percentage of ARI cases	ECYa		No.	ECYa		No.	Percentage of total diarrhoea visits	ECYa		No.	Percentage of total diarrhoea visits	ECYa		No.	CFR (deaths per 1000)
2004–05	3 587 477	1 290 632	0.36		430 700	33.37	0.12		27 103	2.1	0.007		785 336	0.22		296 844	37.8	0.08		17 506	2.2	0.005		244	0.3
2005–06	3 633 687	1 472 562	0.40		468 838	31.8	0.13		23 560	1.6	0.0065		739 915	0.20		248 717	33.6	0.07		10 068	1.4	0.003		82	0.11
2006–07	3 672 650	1 498 359	0.41		464 460	31.0	0.13		18 038	1.2	0.005		680 819	0.18		218 284	32.1	0.06		7 573	1.1	0.002		113	0.17
χ² for linear trend in proportions		18 449.6*			1 706.4*				3 087.5*				12 554.9*			5 182.8*				2 824*				35.6*

* P < 0.01.
ARI, acute respiratory infection; CFR, case fatality rate; ECY, episodes per child per year.
Data obtained from the Ministry of Health and Population.¹⁴

The rate of reported ARI cases was also higher in intervention districts (Table 2). In districts with interventions, 996 369 cases of ARI (55.4%) were reported in a total under-five population of 1 798 668, whereas in the 42 districts without interventions 501 990 cases (26.8%) were reported in a total under-five population of 1 873 982 (χ² = 310 971; P < 0.01). In districts with interventions, pneumonia cases reported as pneumonia and severe pneumonia numbered 282 201 (15.7%) in the under-five population, whereas districts without interventions reported 200 297 cases (10.7%) (χ² = 20 115; P < 0.01). In districts with interventions, the proportion of ARI cases reported as pneumonia (27.6%) was significantly lower than in districts without interventions (37.7%; χ² = 15 802.6; P < 0.01). The proportion of ARI cases reported as severe pneumonia was also lower in districts that received interventions (0.7%) than in non-intervention districts (2.2%; χ² = 6439.5; P < 0.01).

Table 2. Cases of acute respiratory infection and diarrhoea per 100 000 children under 5 years of age in districts in Nepal with and without a community-based programme for the control of diarrhoeal diseases and acute respiratory infections, 2006-2007.

Diagnosis	Intervention districtsa (n = 33)		Non-intervention districtsb (n = 42)
Non-pneumonia ARI	39 705		16 099
Pneumonia	15 304		10 096
Severe pneumonia	386		592
Diarrhoea without dehydration	14 661		10 206
Diarrhoea with some dehydration	6 169		5 727
Diarrhoea with severe dehydration	173		238

ARI, acute respiratory infection.
^a Population under 5 years of age: 1 798 668.
^b Population under 5 years of age: 1 873 982.
Data obtained from the Ministry of Health and Population.¹⁴

During the period from 2004 to 2007, the national pneumonia incidence remained unchanged at around 0.13 episodes per child per year, but the proportion of cases of ARI with pneumonia or severe pneumonia showed a significant trend towards a decrease. The ARI mortality rate remained low and did not show a trend towards a decrease (Table 1).¹⁴

If the programme were expanded to the remaining 42 districts, it could potentially detect 442 372 more cases of ARI and 97 597 more cases of pneumonia per year. This would prevent severe pneumonia in 3860 more children each year. If applied nationwide, the programme would also detect 83 486 more episodes of diarrhoea, including 8283 more episodes of diarrhoea with some dehydration and 1218 more with severe dehydration each year.

Antimicrobial resistance has not emerged in these community-based programmes.¹⁵

Conclusion

The experience in Nepal has shown that community members with minimal prior education can be trained and engaged in their own communities to scale up safe and effective interventions. An important factor in the success of these interventions is their integration into comprehensive community-based programmes for the prevention and control of diarrhoea and respiratory infections in children (Box 1). Engaging CHVs to implement simple strategies to manage cases of diarrhoea and ARI, along with improved nutrition and vaccine coverage, help maintain the focus on problems of national importance.

Box 1. Summary of main lessons learnt.

• Safe, effective and inexpensive interventions are available and can be integrated into community-based programmes to control diarrhoea and acute respiratory infections.

• Community members with minimal prior education can be trained to assess children and provide appropriate treatment or referral to health-care facilities.

• Simplifying and appropriately modifying the training to better meet the needs and use the capacities of community health workers help the programme to succeed.

Success of the programme is ensured by simplifying and modifying the training appropriately so that it meets the needs and uses the capacities of community health workers. Scale-up is best done in a step-wise fashion based on the estimated disease burden in different districts rather than on ease of access. The scaling-up process is facilitated by piloting and monitoring. Outcomes are also improved by the simultaneous use of common human and technical resources at the community level. Supervision and logistical support are essential.

National research data are used to develop policy and are essential to identify and refine effective local tools to prevent and control health problems of national priority. This research-guided shift in emphasis – from externally advocated facility-based strategies to strategies that integrate existing community strengths – is working in Nepal and is likely to be useful in other countries as long as programmes are adapted appropriately for different local contexts. ■