Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Indian Journal of Community Medicine

Evaluation of WHO Diagnostic Algorithm for Reproductive Tract Infections Among Married Women

Author(s): Roochika Ranjan, A.K. Sharma, Geeta Mehta*

Vol. 28, No. 2 (2003-04 - 2003-06)

Deptt, of Community Medicine, *Deptt. of Microbiology, Lady Hardinge Medical College, New Delhi- 110001

Abstract:

Research question: What is the sensitivity and specificity of WHO's syndromic approach in diagnosing Reproductive Tract Infections (RTIs)?

Objective: To test the validity of WHO diagnostic algorithm in diagnosing RTIs among married women.

Study design: Cross-sectional study.

Setting: Primary Health Centre, Palam, New Delhi.

Participants: Married women attending antenatal and gynae clinics.

Sample size: 300 married women.

Statistical analysis: Proportions.

Results: The prevalence of RTIs in married women was 37.0% by syndromic approach based on symptoms, 51.7% by clinical examination and 36.7% by microbiological laboratory investigations. The sensitivity and specificity of syndromic approach to diagnose any RTI was 53.6% and 72.6% respectively while clinical examination had 68.2% sensitivity and 60.5% specificity. Overall clinical examination had relatively high sensitivity but low specificity. For trichomoniasis and bacterial vaginosis the clinical examination had low sensitivity but a high specificity.

Conclusions: WHO syndromic approach based on symptoms had a low sensitivity in diagnosing RTIs among women. Sensitivity increased when clinical examination was used for the diagnosis of these infections. In the absence of microbiological laboratory facilities, syndromic approach should be supplemented with clinical examination for diagnosing RTIs in women to avoid over-treatment of women.

Keywords: Reproductive tract infections, WHO syndromic approach, Sensitivity, Specificity

Introduction:

Reproductive tract infections (RTIs) are a major public health problem among women especially in developing countries. Women tend to suffer more because of the synergistic effects of infection, malnutrition and reproduction. Failure to provide effective treatment for these infections can lead to continued spread of disease, high rate of complications and enhanced rate of sexual transmission of HIV/AIDS1.

WHO has recommended a syndromic management of reproductive tract infections, which is based on symptoms and clinical signs2. Through this approach a health worker at the most peripheral level without using laboratory support, can diagnose these infections and accordingly prescribe treatment or advise referral of the patient3,4.

This study aims to evaluate the ability of syndromic approach as given by WHO to diagnose various reproductive tract infections among married women.

Material and Methods:

The study was conducted at Primary Health Centre Palam, which is a rural field practice area of lady Hardinge Medical College, New Delhi. 300 currently married women were included in the study of which 200 were from antenatal clinic and 100 from gynae clinic. It was an observational cross-sectional study. A pre-tested, semi-structured questionnaire was administered which included information about demographic, socio-economic status, obstetric and gynaecological history.

After history taking, general physical examination, per abdomen, per speculum and per vaginum examinations were done. During per speculum examination, vaginal and endocervical swabs were taken for direct examination and culture. All the specimens were then transported to the Microbiology Department for processing as soon as possible. The specimens were processed by standard techniques5.

The completed questionnaires were entered in the computer for analysis after checking. The data was analysed and tabulated using the SPSS software (version 7.5). The diagnosis of various reproductive tract infections by syndromic approach based on symptoms and clinical examination was evaluated against laboratory diagnosis as the gold standard6,7.

Observations: About 72.0% of the women in the study population were in the age group of 20-29 years while 8.7% of the women were in their teens. Majority (79.0%) of the study population were Hindus followed by Sikhs (10.0%), Muslims (8.3%) and Christians (2.7%). 39.3% of the study population was just literate or were educated upto primary level and 27.3% were illiterate. About 45% of the women in the study population belonged to upper middle socioeconomic class, while 28.3% were from lower middle and 25.7% belonged to upper lower class. About 60.0% of the women in the study population belonged to nuclear families, 16.0% to joint families and 24.3% to extended families.

Table I: Reproductive tract infections identified by different approaches.

Approach Prevelance of RTI's
n=300
Syndromic approach based on symptoms 37.0%
Clinical examination 57.7%
Laboratory Diagnosis 36.7%

The prevalence of RTIs was 37.0% by syndromic approach based on symptoms, 51.7% by clinical examination and 36.7% by laboratory investigations.

Table II: Reproductive tract infections in married women by syndromic approach, based on symptoms.

Symptom Total
(n=300)
%
Vaginal discharge 87 29.0
Lower abdomen pain 61 20.3
Genital ulcer 0 0
Inguinal Bubo 0 0
No symptom 189 63.0

Vaginal discharge and/or lower abdominal pain were the only two symptoms of reproductive tract infections present in women. These were present in 29.0% and 20.3% women respectively. Neither genital ulcer nor inguinal bubo was present in any of the subjects.

Table III: Reproductive tract infections in married women by clinical examination.

Reproductive tract infectioin Total (N=300) %
Candida 84 28.0
Trichomonas/Bacteriological vaginosis 51 17.0
Chlamydia/Gonnorrhoea 15 5.0
Pelvic Inflammatory Disease 16 5.3
No RTI 145 48.3

Based on clinical examination, the prevalence of reproductive tract infections was about 52.0%. Candidiasis was the most common (28.0%) infection followed by trichomoniasis/bacterial vaginosis (17.0% )

Table IV: Reproductive tract infections in married women by laboratory investigations.

Reproductive Tract INfection Total (N=300) %
Candidiasis 79 26.3
Vaginitis 54 18.0
Trichomoniasis 47 15.7
BActerial Vaginosis 43 14.3
PID 18 6.0
Cervicitis 14 4.7
Gonorrhoea 4 1.3
Syphilis 3 1.0
No RTI 190 63.3

The overall prevalence of reproductive tract infections, by laboratory investigations was 36.7%. Prevalence of Candidiasis was highest (26.3%) among all infections diagnosed. It was followed by vaginitis (18.0%), trichomoniasis (15.7%) and bacterial vaginosis (14.3%).

Table V: Sensitivity, specificity and predictive values of syndromic approach and clinical examination in diagnosing RTIs.

Approach Sensitivity Specificity PPV NPV
Syndromic approach based on symptoms 53.6% 72.6% 53.1% 73.0%
Clinical examination 68.2% 60.5% 50.0% 76.7%

The sensitivity and specificity of syndromic approach to diagnose RTI was 53.6% and 72.6% respectively, while clinical examination had 68.2% sensitivity and 60.5% specificity. An improvement in sensitivity was observed when RTIs were diagnosed with the help of clinical examination (68.2%), however, specificity decreased (60.5%) as compared to (72.6%) diagnosis by syndromic approach.

Syndromic approach based on symptoms and clinical examination both had a low positive predictive value (PPV) in diagnosing these infections and would result in large number of women receiving unnecessary treatment. For trichomoniasis and bacterial vaginosis the clinical examination had a low sensitivity but a high specificity.

Table VI: Sensitivity, specificity and predictive values of clinical examination in diagnosing different RTIs.

Reproductive Tract Infection Sensitivity Specificity PPV NPV
Candidiasis 56.2% 82.7% 54.2% 83.9%
Trichomoniasis 21.3% 76.3% 19.6% 78.1%
Bacteriological vaginosis 32.5% 85.6% 27.4% 88.3%

Clinical examination based approach had a low PPV in diagnosing trichomoniasis and bacterial vaginosis which would result in a large number of women being over treated for these infections.

Discussion:

Present study showed that the WHO syndromic approach based on symptoms had a low sensitivity (53.6%) in diagnosing RTIs among women. Clinical examination was observed to be more sensitive (68.2%) in diagnosing these infections among women. Both the approaches had a low PPV which would lead to over treatment of women who did not have the disease, taking laboratory diagnosis as the gold standard.

Mayaud et al8 (1994) evaluated the performance of WHO algorithm which showed a sensitivity of 10.2%, specificity of 92% and positive predictive value of 9.8% in the detection of N. gonorrhoea and/or C. trachomatis cervical infection. Bourgeois et al9 (1994-95) evaluated an algorithm based on symptoms like pelvic pain, vaginal discharge and characteristics of vaginal discharge. The midwives recorded the sensitivity and specificity of the algorithm as 73.3% and 54.8% respectively as compared to 76.7% sensitivity and 50.6% specificity recorded by physicians. Kapiga et al10 (1995) observed that a screening strategy for cervicitis based on symptoms had a sensitivity of 29.7%, specificity of 84.1% and a positive predictive value of 15.9%. The corresponding figures for algorithm based on clinical signs were 20.3%, 90.2% and 15.6% respectively. Vishwanath et al11 (1999) found that the algorithm based on risk assessment and speculum examination was not helpful in predicting cervical infections associated with C. trachomatis (sensitivity 5% and PPV 9%). This algorithm was sensitive (95%) though not specific (22%) in diagnosing bacterial vaginosis or trichomoniasis and over treatment was a problem (PPV 38%). The sensitivity, specificity and PPV of this algorithm for the diagnosis of candidiasis were 48%, 98% and 88% respectively. A study by Nandan et al12 (1999-2000) observed an improvement in the sensitivity (81.8%) and a predictive accuracy (74.1%) in the diagnosis of RTI by clinical examination as compared to the syndromic approach.

Conclusions:

WHO syndromic approach based on symptoms had a low sensitivity in diagnosing RTIs among women. Sensitivity increased when clinical examination was used for the diagnosis of these infections. Clinical examination had a low sensitivity and a high specificity particularly in diagnosing trichomoniasis and bacterial vaginosis. As PPV was also low for these infections it would lead to unnecessary treatment of women for these infections in addition to the escalation of the cost of treatment. Therefore, when laboratory facilities are not available for diagnosing RTIs, syndromic approach should be used along with clinical examination for the diagnosis of RTIs.

References:

  1. The World Bank. World Development Report: Investing in health, New York. Oxford .University Press 1993.
  2. World Health Organization. Flow charts on management of sexually transmitted diseases, WHO, Regional Office for South East Asia, New Delhi, 2000.
  3. Government of India, Ministry of Health and Family Welfare. Simplified RTI and STI treatment guidelines. National AIDS Control Organization, New Delhi, 1999.
  4. World Health Organization. Management of patients with sexually transmitted diseases. Report of a WHO Study Group. WHO Technical Report Series 810, WHO, Geneva 1991.
  5. Collee JG, Marr W. Culture of bacteria. In: Collee JG, Fraser AG, Marmion BP, Simons A, eds., Mackie and Me Cartney's Practical Medical Microbiology, 14th Edition. Longman Publishers Ltd., 1996: 113-30.
  6. World Health Organization. Management of sexually transmitted diseases at district and PHC levels. Regional Publication, WHO, Regional Office for South-East Asia, New Delhi, 1999.
  7. National AIDS Control Organization. Reference manual for laboratory workers. Diagnosis of sexually transmitted diseases. New Delhi, 1994,
  8. Mayaud P, Uleidi E, Cornelissen J et al. Risk scores to detect cervical infections in urban antenatal clinic attendees in Mwanza, Tanzania. Sexually Transmitted Infections 1998; 741 Supp (ii): S139-S46.
  9. Bourgeois A, Henzel D, Dibanga G et al. Prospective evaluation of a flowchart using a risk assessment for the diagnosis of STDs in primary health care centres in Libreville, Gabon. Sexually Transmitted Infections 1998; 74(Suppl 1): S128-S131.
  10. Kapiga SH, Vuylsteke B, Lyamuya EF et al. Evaluation of sexually transmitted diseases diagnostic algorithms and family planning clients in Dar es Salaam, Tanzania. Sexually Transmitted Infections 1998;74(Suppl 1): S132-S138.
  11. Vishwanath S, Talwar V, Prasad R. Syndromic management of vaginal discharge among women in a reproductive health clinic in India. Sexually Transmitted Infections 2000; 76: 303-6.
  12. Nandan D, Gupta YP, Krishnan V et al. Reproductive tract infections in women of reproductive age group in Sitapur and Shahjahanpur districts of UttarPradesh. Indian Journal of Public Health 2001; 45(1): 8-13.
Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica