Elsevier

Social Science & Medicine

Volume 63, Issue 6, September 2006, Pages 1648-1660
Social Science & Medicine

Global review of health care surveys using lot quality assurance sampling (LQAS), 1984–2004

https://doi.org/10.1016/j.socscimed.2006.04.011Get rights and content

Abstract

We conducted a global review on the use of lot quality assurance sampling (LQAS) to assess health care services, health behaviors, and disease burden. Publications and reports on LQAS surveys were sought from Medline and five other electronic databases; the World Health Organization; the World Bank; governments, nongovernmental organizations, and individual scientists. We identified a total of 805 LQAS surveys conducted by different management groups during January 1984 through December 2004. There was a striking increase in the annual number of LQAS surveys conducted in 2000–2004 (128/year) compared with 1984–1999 (10/year). Surveys were conducted in 55 countries, and in 12 of these countries there were 10 or more LQAS surveys. Geographically, 317 surveys (39.4%) were conducted in Africa, 197 (28.5%) in the Americas, 115 (14.3%) in the Eastern Mediterranean, 114 (14.2%) in South-East Asia, 48 (6.0%) in Europe, and 14 (1.8%) in the Western Pacific. Health care parameters varied, and some surveys assessed more than one parameter. There were 320 surveys about risk factors for HIV/AIDS/sexually transmitted infections; 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 surveys on other diseases. LQAS is a practical field method which increasingly is being applied in assessment of preventive and curative health services, and may offer new research opportunities to social scientists. When LQAS data are collected recurrently at multiple time points, they can be used to measure the spatial variation in behavior change. Such data provide insight into understanding relationships between various investments in social, human, and physical capital, and into the effectiveness of different public health strategies in achieving particular behavioral outcomes.

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)

  • G. Carrasco

    Lot quality acceptance sampling: A new tool for a new way to monitor quality

    Revista de Calidad Asistancial

    (2000)
  • Investigation of LQAS methodology in the assessment of malnutrition: Preliminary report

    (2003)
  • M. Deitchler et al.

    Report on the field testing of three LQAS designs to assess the prevalence of acute malnutrition

    (2004)
  • S. Deshevoi

    Lot quality coverage survey: National measles–rubella immunization campaign, Bishkek, Kyrgyzstan

    (2002)
  • H.F. Dodge et al.

    Sampling inspection tables. Single and double sampling

    (1959)
  • F. Gasse

    Trip report, Indonesia

    (1996)
  • M.D. Gupte et al.

    Application of lot quality assurance sampling for leprosy elimination monitoring—examination of some critical factors

    International Journal of Epidemiology

    (2004)
  • R.H. Henderson et al.

    Cluster sampling to assess immunization coverage: A review of experience with a simplified sampling method

    Bulletin of the World Health Organization

    (1982)
  • Hoshaw-Woodard, S. (2001). Description and comparison of the methods of cluster sampling and lot quality assurance...
  • Y.J.F. Hutin et al.

    Trypanosoma brucei gambiense trypanosomiasis in Terego county, northern Uganda, 1996: A lot quality assurance sampling survey

    American Journal of Tropical Medicine and Hygiene

    (2004)
  • Evaluation of immunization services in Kosovo using the lot quality technique

    (2002)
  • C.O. Khoromana et al.

    In vivo efficacy of chloroquine treatment for Plasmodium falciparum in Malawian children under five years of age

    Journal of Tropical Medicine and Hygiene

    (1986)
  • A. Lakshmi

    A statistical approach to monitor ongoing intervention for control of lymphatic filariasis

    Journal of Communicable Diseases

    (2000)
  • C.F. Lanata et al.

    Lot quality assurance sampling techniques in health surveys in developing countries: Advantages and current constraints

    World Health Statistics Quarterly

    (1991)
  • S. Lemeshow et al.

    Adequacy of sample size in health studies

    (1990)
  • S. Lemeshow et al.

    Quality assurance sampling for evaluating health parameters in developing countries

    Survey Methodology Statistics Canada

    (1989)
  • S. Lemeshow et al.

    Lot quality assurance sampling: Single- and double-sampling plans

    World Health Statistics Quarterly

    (1991)
  • S.K. Lwanga et al.

    Immunization coverage surveys: Methodological studies in Indonesia

    Bulletin of the World Health Organization

    (1987)
  • S.K. Lwanga et al.

    Sample size determination in health studies: A practical manual

    (1991)
  • Report of a technical meeting on the use of lot quality assurance sampling (LQAS) in polio eradication programs

    (1998)
  • J. Mukaire et al.

    Assessment of HIV/AIDS related knowledge, practices and coverage in 19 districts supported by the Uganda AIDS Control Project

    (2003)
  • B.N. Murthy et al.

    Lot quality assurance sampling LQAS for monitoring leprosy elimination in an endemic district in Tamilnadu

    Indian Journal of Leprosy

    (2001)
  • M. Myatt

    Guidelines for the rapid assessment of trachoma prevalence using ASTRA surveys

    (2003)
  • M. Myatt et al.

    Field trial of the applicability of lot quality assurance sampling survey method for rapid assessment of prevalence of active trachoma

    Bulletin of the World Health Organization

    (2003)
  • B. Nsimba et al.

    Efficacy of sulfadoxine/pyrimethamine in the treatment of uncomplicated Plasmodium falciparum malaria in Republic of Congo

    American Journal of Tropical Medicine and Hygiene

    (2004)
  • L.P. Rabarijaona et al.

    Use of lot quality assurance sampling in the identification of Schistosoma haematobium endemic communities in Madagascar

    Archives de l’Institut Pasteur de Madagascar

    (2001)
  • Cited by (108)

    • An Integrated Infant and Young Child Feeding and Small-Quantity Lipid-based Nutrient Supplementation Program Is Associated with Improved Gross Motor and Communication Scores of Children 6-18 Months in the Democratic Republic of Congo

      2020, Journal of Pediatrics
      Citation Excerpt :

      The enhanced IYCF program was initiated in the intervention zone in September 2012, with the first SQ-LNS distributions occurring in May 2013. After program launch in 2012, 3 rounds of lot quality assurance survey33 were carried out in all 13 health areas or posts of the Kasenga (intervention) health zone during February 2013, July 2013, and February 2014. Health areas or posts were graded on 21 program fidelity benchmark questions ranging from SQ-LNS stocks, staff training, health worker activities, distribution of SQ-LNS, and group counseling.

    • Using rapid-cycle evaluation to improve program design and delivery

      2023, The Oxford Handbook of Program Design and Implementation Evaluation
    View all citing articles on Scopus
    View full text