Global review of health care surveys using lot quality assurance sampling (LQAS), 1984–2004
Introduction
Effective management of decentralized health systems requires up-to-date information at the local level. Health program managers need to know which communities are meeting particular targets and goals, and which are not. Interest in applying lot quality assurance sampling (LQAS) to health assessments has been growing since the mid-1980s (Khoromana, Campbell, Wirima, & Heymann, 1986; Lemeshow, Hosmer, Lwanga, & Klar, 1990; Lemeshow & Stroh, 1989; Lwanga & Abiprojo, 1987; Smith, 1989; Wolff & Black, 1989). The LQAS method was originally developed as a quality control technique for goods produced in factories, but its sampling concepts have universal applicability. In 1991, a World Health Organization (WHO) consultation on epidemiological and statistical methods for rapid health assessment concluded that LQAS was one of the more practical methods, and encouraged its further development to monitor health programs (Anker, 1991; Lanata & Black, 1991; Lemeshow & Taber, 1991).
The growing interest in using the LQAS method was captured in a 1997 WHO review of 34 LQAS applications assessing immunization coverage, antenatal care, use of oral rehydration therapy, growth monitoring, family planning, disease incidence, and the technical skills and knowledge of health workers (Robertson et al., 1997). The LQAS method has also been used to assess the accuracy of health records, outreach of community health workers, and health worker training programs (Valadez, 1991; Valadez, Brown, Vargas, & Morley, 1996; Valadez, Transgrud, Mbugua, & Smith, 1997). In Armenia, Malawi, and Nicaragua, networks of nongovernmental organizations have used LQAS to track national disaster relief and reproductive health programs (Valadez, Hage, & Vargas, 2005; Valadez, Leburg, Mirzoyan, Matevosyan, & Veloso, 2001; Vargas, Valadez, Manda, & Mobley, 2003). In Kenya and Uganda, it has been used to assess the performance of HIV/AIDS control programs at the district and subdistrict levels as well as nationally (Mukaire, Kitui, Ssekamatte-Ssebuliba, & Valadez, 2003). There are now several examples applying LQAS for ongoing supervision, the most well-documented from a maternal and child health project in rural Nepal (Valadez & Devkota, 2002).
The burgeoning use of LQAS by managers may offer new research opportunities to social scientists and the readership of this journal. When LQAS data are collected recurrently at multiple time points, the data can be used to measure the spatial variation in behavior change in the different management units comprising a project catchment area. Such data provide insight into understanding relationships between various investments in social, human, and physical capital, and behavior they are intended to change (Valadez et al., 2005). Such work, although in its infancy, could influence our understanding of which public health strategies may best achieve particular behavioral outcomes.
The present review has identified 805 LQAS surveys on health, indicating widespread uptake of this method. This paper reviews the general principles of the LQAS method, documents LQAS survey training materials available, and provides a descriptive survey of its uses in health assessments during 1984–2004.
Section snippets
General principles of the LQAS method
During the 1980s, health system evaluators explored the applications of industrial quality control methods to assess health worker performance (Stroh, 1983; Reinke, 1988; Valadez, Villegas, & Bixby, 1986). LQAS was originally developed in the 1920s to control the quality of industrially produced goods (Dodge & Romig, 1959). The principle is that a supervisor inspects a small representative sample of a recently manufactured lot of goods from a production unit, such as an assembly line or
LQAS training materials
In 1991 WHO published general guidelines on use of the LQAS method in health studies (Lwanga & Lemeshow, 1991). In the same year, Harvard University published a book on implementation of the LQAS method for assessing child survival programmes (Valadez, 1991). A general manual on LQAS was developed as part of a public health training program in Spain (Saturno, 2000b). More recently, a detailed training manual on using the LQAS technique to assess community health care programmes was published (
Literature search
Studies of “lot quality assurance sampling”, “lot quality acceptance sampling”, “lot quality”, “LQAS”, “LQA”, “LQ”, and “health” were identified by conducting online literature searches. There were no restrictions on language of publication or date of publication. We searched the following databases: Medline, African Index Medicus, CAB Health Database, Index Medicus for the Eastern Mediterranean Region, Latin American and Caribbean Health Science Information, and Yahoo.
We also reviewed reports
Results: global findings
Altogether 805 LQAS surveys were identified from a total of 123 data sources. Language of the source document was: English 87 (70.7%), Spanish 16 (13.0%), English and French 8 (6.5%), French 6 (4.9%), Portuguese 3, Chinese 2, Turkish 1. Data sources for the 805 surveys were: report 649 (80.6%), journal 129 (16.0%), book 16, personal communication 7, abstract 3, thesis 1. Formal publication of the survey findings (book, journal, abstract) occurred for 148 (18.4%) surveys as of December 2004.
Results by survey parameter
Health parameters assessed in the surveys varied over time, and some surveys assessed more than 1 health parameter (Table 2). There were 320 surveys concerning risk factors for HIV/AIDS/STI; 266 surveys on immunization coverage, 240 surveys post disasters, 224 surveys on women's health, 142 surveys on growth and nutrition, 136 surveys on diarrheal disease control, and 88 surveys on quality management. LQAS surveys to assess disease burden included 23 neonatal tetanus mortality surveys and 12
LQAS surveys on HIV/AIDS/STI
We identified 320 surveys on HIV/AIDS/STI. Several countries conducted many surveys: Uganda 180 surveys (Mukaire et al., 2003), Malawi 63 surveys (Vargas et al., 2003), Nicaragua 34 surveys (Campos, Valadez, & Vargas Vargas, 2002). A total of 319 surveys assessed sexual behavior and risk factors for HIV and other sexually transmitted infections. The target group was adult women for 133 surveys (41.7%), adult men for 64 surveys (20.1%), youths for 62 surveys (19.4%), AIDS orphans for 30 surveys
LQAS surveys on immunization coverage
We identified 266 LQAS surveys that measured immunization coverage. Coverage of women with tetanus toxoid was assessed in 149 surveys. The survey target population was mothers of infants in 91 surveys, mothers of children 12–23 months in 32 surveys, mothers of children in other age groups in 9 surveys, and women of childbearing age in 17 surveys. One hundred twelve surveys assessed immunization coverage in children, usually at 12–23 months of age. Five immunization coverage surveys targeted
LQAS surveys in disaster settings
We identified 240 LQAS surveys conducted following disasters. In Nicaragua, 123 surveys were conducted following a devastating hurricane which destroyed much of the local infrastructure (Campos et al., 2002). In Turkey, an LQAS survey assessed immunization coverage after an earthquake (S. Sener, 2002, personal communication). LQAS surveys were conducted in post-war settings in Afghanistan (Ameli, Ickx, & Huger, 2004), Serbia and Montenegro (Institute of Public Health, 2002), and Eritrea (WHO,
LQAS surveys on women's health
We identified 224 LQAS surveys concerning women's health: 91 were conducted in Nicaragua and 54 in Afghanistan. In most of these surveys women were asked about receipt of tetanus toxoid, prenatal vitamin A supplementation, and breastfeeding of their infants. Eighty-three surveys assessed family planning knowledge and use by women, and 38 of these also examined HIV/AIDS/STI risk factors. Twenty-three surveys on women's health asked men about their knowledge of safe motherhood. Three surveys
LQAS surveys on growth and nutrition
We identified 142 LQAS surveys on growth and nutrition: 75 surveys on delivery of vitamin A to children or pregnant women and 62 surveys on breastfeeding practices. Two surveys assessed acute malnutrition in Ethiopia (Deitchler, Valadez, Egge, & Fernandez, 2004). A manual on assessing malnutrition using LQAS is under development, and this should prove useful in future emergencies (Center for Biostatistics, 2003). Three growth and nutrition surveys were in the area of quality assurance,
LQAS surveys on diarrheal disease control
There were 136 LQAS surveys on diarrheal disease control. It is not surprising to find many LQAS surveys on diarrheal disease control, as the first LQAS manual for health concerned surveys on uptake of oral rehydration therapy (Wolff & Black, 1989). The majority of surveys interviewed mothers on their use of oral rehydration therapy or other treatments for diarrhea in young children. The target age group was 0–11 months in 44 surveys, 12–23 months in 48 surveys, 0–23 months in 31 surveys, and
LQAS surveys on quality management
There were 88 LQAS surveys on quality management. In 40 surveys the performance of health care workers was monitored by direct observation of their treatment of patients. Surveys were directed at health care personnel ranging from community health care workers to nurses and physicians. Usually these were conducted following a training course concerning specific skills, e.g., family planning or immunization (Valadez, 1991; Valadez, Vargas, & Diprete, 1990; Valadez, Weld, & Vargas, 1995). In
LQAS surveys on neonatal tetanus mortality
WHO has developed a protocol that uses a combination of LQAS and cluster survey methods to assess whether the incidence of neonatal tetanus is below the elimination threshold of one death due to neonatal tetanus per 1000 live births (Stroh & Birmingham, 2002). The survey requires a sample size of 1000–3000 live births, includes a two-stage sampling scheme, and uses verbal autopsy methods to assess the cause of neonatal deaths. Field tests of the new method were conducted in Bangladesh (Stroh,
LQAS surveys to assess other diseases
We identified 12 surveys of other diseases using the LQAS method. Broadly there were two types of surveys: (i) assessment of disease prevalence and (ii) assessment of drug treatments. These surveys were complex to implement, as they required clinical examination of patients, collection of patient specimens, and subsequent laboratory analyses.
There were six LQAS surveys on disease prevalence. Two studies in India assessed leprosy prevalence-based on clinical examination and skin smears for
Discussion
Our global reviewed identified 805 LQAS surveys on health carried out during 1984–2004. Initially, uptake of this survey method was low, but there was a greater than 12-fold increase in the annual number of surveys conducted in 2000–2004. Until recently most LQAS surveys were conducted on an exploratory basis, usually by research institutions. An impediment in applying the method was difficult-to-understand statistical explanations that were not helpful to public health professionals interested
Acknowledgements
This review was carried out with support from the World Bank and the World Health Organization. Thanks to Yoko Shimada for help in identifying and cataloging surveys, Ravi Cherukerupalli for assistance in database management, and Sandra Garnier for help with graphics. Appreciation to Martha Anker, Pedro Saturno, and Jos Vandelaer for comments on the manuscript. Special thanks to Susan Stout for her invaluable support.
References (72)
- et al.
Interest of LQAS method in a survey of HTLV-I infection in Benin (West Africa)
Journal of Clinical Epidemiology
(2002) - et al.
Validity of lot quality assurance sampling to optimize falciparum malaria surveys in low-transmission areas
Transactions of the Royal Society of Tropical Medicine and Hygiene
(2001) - et al.
Understanding the relationship of maternal health behavior change and intervention strategies in a Nicaraguan NGO network
Social Science & Medicine
(2005) - et al.
In vivo drug resistance of falciparum malaria in mining areas of Venezuela
Tropical Medicine and International Health
(2002) A system for monitoring therapeutic efficacy of antimalarial drugs against Plasmodium falciparum infections in the Sudan
(2000)- et al.
Rural expansions of Afghanistan's community-based health care program: NGO grantees’ baseline household survey
(2004) Epidemiological and statistical methods for rapid health assessment: Introduction
World Health Statistics Quarterly
(1991)External quality assessment for AFB smear microscopy
(2002)- Bremer, V., Leitmeyer, K., Dabbagh, A., Brou, A., Doua, K., Koumare, B., et al. (2002). Evaluation of yellow fever...
- et al.
The NicaSalud Network: Restoring community health activities in Nicaragua after Hurricane Mitch, a final evaluation
(2002)