A novel system to obtain addresses of out-patients--assessment in routine clinic practice in Madras, S. India.

A novel method of obtaining accurate home addresses from out-patients was introduced as a routine procedure in 6 chest clinics of Madras City, following highly satisfactory results under study conditions. In this method, the patient is given a card (the address card), and asked to get his exact address entered on it by any knowledgeable person of his choice such as a landlord or neighbour. An assessment of the system was undertaken after it had been in operation for about 8 months. A complete and legible address was available for 82% of 3956 patients, the range in the 6 clinics being 74% to 91%. The main causes for failure were: not giving address card to patient (7%), patient not reattending the clinic (6%), and patient reattending but not returning the address card (3%). Corrective measures have now been introduced, and a re-assessment will be undertaken in due course.

Study of a policy to minimise the prescription of medicaments at first attendance at chest clinics.

In chest clinics in Madras, south India, medicaments are prescribed to many patients at the first clinic attendance, whether necessary on medical grounds or not, in the belief that this practice will increase the likelihood of the patients subsequently reattending the clinic. This study of 2608 patients in four chest clinics showed that the proportion prescribed medicaments ranged from 50% to 75%. Subsequently, a modified policy of prescribing medicaments only when they were medically essential was investigated on 956 patients in the largest of these clinics. The policy was found to be practicable, and it did not have any adverse consequences such as an increased rate of default or an unacceptable level of patient dissatisfaction. The advantages of the new policy are savings in money, man-power, and time and the potential for a reduction in the incidence of side-effects.

A randomised study of two policies for managing default in out-patients collecting supplies of drugs for pulmonary tuberculosis in a large city in South India.

A randomized controlled study was undertaken to compare 2 policies of default management in out-patients with smear-negative pulmonary tuberculosis attending a large chest clinic in Madras city. All the patients were due to collect monthly supplies of drugs for a year, for daily self-administration at home. In the routine (R) policy, if a patient failed to collect the drug supply on a due date, a reminder letter was posted on the fourth day and, if necessary, a health visitor visited the home a week later. In the intensive (I) policy, a health visitor visited the home on the 4th day and, if necessary, a week later and at 1 and at 2 months. The main analyses concern 150 patients (75 R, 75 I), of whom 16 R and 15 I patients had a positive culture. A total of 29 patients (11 R, 18 I) did not default at any time. For the remaining 64 R and 57 I patients, the mean numbers of defaults were 3.0 and 2.3, and the mean numbers of defaulter retrieval actions were 4.3 and 3.8, respectively. The home visit as the first action (I series) was successful in retrieving defaulters on 65% of 132 occasions, while the reminder letter (R series) was successful in 56% of 193 occasions (P = 0.1). Following the second action, which was a home visit in both the series, these proportions became 80% and 84%, respectively. in the I series, 22 third and 18 fourth actions were taken, but the patient was retrieved in only 4 and 0 instances respectively. The mean number of drug collections during the year was significantly higher in the I series (9.8) than in the R series (8.6). Finally, the proportions of patients who made 12 collections in a 15-month period, a satisfactory target under Indian Programme conditions, were 69% and 52%, respectively (P = 0.07).

A study of the accuracy, and factors influencing accuracy, of home addresses of patients obtained by registry clerks and address cards in four large towns in South India.

In 4 large towns in South India with illiteracy levels of 26% to 40%, the efficiency of registry clerks in eliciting the home addresses of 1338 out-patients was assessed, by verifying receipt of a letter posted to the patients. The efficiency was found to be very poor, namely 66%. Moreover, the accuracy of address was substantially poorer for illiterate patients and for patients living for relatively short durations at their present address. Our innovation, the address card, on which the home address was recorded by a knowledgeable literate person of the patient's choice, was returned by 98% of the patients, and the addresses were found accurate in 84%; the findings were similar in the 4 towns and were unaffected by any patient characteristic. The substantially better results with the address card were found in both illiterate and literate patients. These findings establish the address card as a simple, inexpensive and efficient device for obtaining accurate addresses.

Efficiency of address cards, experienced health visitors and motivated registry clerks in obtaining the home address of urban patients in South India.

The address card, a card on which the patient's home address is asked to be recorded by the local postman, or by a knowledgeable and literate neighbour, relative or friend, was investigated for acceptability and efficiency in 4 tuberculosis out-patient clinics, in an urban community with substantial levels of illiteracy in Madras City. In the 4 clinics combined, 96% of the patients who reattended returned the completed card. Letters posted to the address on the cared were received by 85% of 419 patients, while 5% were returned by the post office as undelivered and a further 4% were, in all probability, not delivered; no information was available about the remaining 6%. A formal comparison in 392 of the above patients demonstrated the address card method to be significantly more efficient than interrogation by experienced health visitors. A retrospective comparison suggested that the efficiency of experienced health visitors was slightly better than that of highly motivated registry clerks, the proportions of letters received being 72% and 65% respectively.

An investigation of the accuracy of the home address given by patients in an urban community in South India.

Studies were undertaken in three tuberculosis clinics in Madras, a large Indian city with a good civic organization, to assess the accuracy of address recorded routinely by registry clerks at the patient's first clinic attendance. The accuracy was poor, with 20% to 30% of the letters posted not reaching the patients. It was appreciably improved, by 10% to 20%, by supplementing the clerk's efforts with questioning by a motivated, experienced health visitor. An address card, a card on which the patient's address was recorded by the local postman or a literate neighbour, relative or friend, was returned by 90% to 94% of the patients, and the accuracy of addresses was found to be at least as good as that obtained with the health visitor. Even when all three sources of information were considered, the patient's home could not be traced in 3% of cases and was found with difficulty in 4%.

Studies of immediate adverse reactions to different doses of a slow-release preparation of isoniazid.

'Double-blind' studies were carried out to assess the incidence of immediate adverse reactions to different doses of a slow-release preparation of isoniazid (matrix isoniazid). Individual doses of 30 mg/kg matrix isoniazid were well-tolerated but higher doses resulted in giddiness, the incidence being dose-related. The giddiness was characterized by a late onset and was usually present even at 24 hours. A few patients complained of gastro-intestinal symptoms. It is concluded that matrix isoniazid can be given to Madras patients in doses of 30-40 mg/kg without risk of an undue incidence of immediate adverse reactions.