H.—3l.

used per one hundred patient-staff days was in one case 69 lb. and in the other 94-9 lb. The hospital with the lower rate showed, however, an increase in the consumption of eggs 5-3 dozen against 3-5 dozen. Milk again was lower in the first case, the figures being 108-7 pints and 123-9 pints respectively. In investigating the consumption of foodstuff quite erroneous conclusions may be drawn if comparisons are made on individual items. When, however, in making inspection, excessive quantity consumption is encountered, attention is drawn to the fact, and it is pleasing to note immediate steps are taken by the hospital authorities to investigate and apply a remedy. A few examples taken from the heading " Surgery and Dispensary " issues will show a similar variation in quantity consumed. Note.—The unit used here is " one hundred patient days " (the consumption of these articles not being affected by the number of staff). Cotton wool varies from 0-85 lb. to 4-4 lb. Gauze varies from 8-6 yd. to 72-3 yd. Spirits —Vini, methylated varies from 2-3 lb. to 11-6 lb. Hydrogen peroxide varies from 0-1 lb. to 1-7 lb. In these instances due weight must be given to the proportion of surgical work done. It is evident, therefore, that in order to secure economy, consumption within the hospital is just as important as the price paid by the hospital. As variations also occur in the prices paid for the same articles* it may be said that this constitutes a strong argument in favour of a combined buying scheme or schemes for our hospitals ; but before such an object could be secured a system of standardization would have to be evolved. This presents considerable difficulties not so much in food lines as in, surgery and dispensary supplies. It appears to me that group buying—that is, the grouping of hospitals geographically in touch —would result in savings just as great as the setting-up of one central buying authority. The sum of my remarks is that for efficient and economical administration not only must every effort be made to secure that, the best prices are obtained, but equal efforts must be devoted to ensuring that the quantity consumed is also within reason. If these two factors are always kept well in the forefront, then an administrator will find that the problem of " waste " is largely solved. It will be obvious that to secure this desirable result the whole hospital staff must be working as a co-ordinated team and that each group or in some instances each individual should be kept in touch with the activities of all other groups or individuals by the issue of frequent comparative returns, augmented by free discussions by administrative heads at regular conferences. In addition to the investigation of actual costs of maintenance and treatment in the various institutions, the need arises from time to time for investigating the volume of clinical work done. The direction in which such investigation should be pursued is often indicated by comparing the proportion of population in each district given inpatient treatment. Here again due weight must be given to various factors such as (1) the number of patients from outside districts seeking treatment, and (2) the facilities available for private treatment in the areas concerned. The figures given below take no account of these factors. It can be taken for granted, however, that comparable private facilities are available in most of the larger centres. The figures quoted refer only to " general beds " and as far as possible exclude admissions to tuberculosis wards, chronic wards, and infectious disease and maternity wards, but include general beds in subsidiary hospitals. Hospital District. Proportion of each 1 000 of Population r treated as Inpatients, 1930-31. Auckland .. .. .. .. .. .. 36-45 Wellington .. .. .. .. .. 46-47 North Canterbury .. .. .. .. .. 42-07 Otago .. .. .. .. .. .. 41-04 Waikato . . .. .. .. .. .. 42-14 Taranaki .. . . .. .. >.. .. 50-44 Wanganui .. .. .. .. .. 48-02 Palmerston North .. .. .. .. .. 46-91 South Canterbury .. .. .. .. .. 47-17 Southland .. .. .. .. .. 36-21 Nelson .. .. .. .. .. .. 38-90 Buller .. .. .. .. .. .. 66-86 Grey .. .. .. .. .. ..87-70 Ashburton .. .. .. .. .. 52-46 Dannevirke .. .. .. . . .. 46-97 Waipawa . . .. .. .. .. .. 81-75 Waihi .. .. .. .. .. .. 129-87 It will be noticed that the proportion varies from 36-21 per 1,000 in Southland to 129-87 in Waihi. The proportions for North Canterbury, Waikato, and Otago are possibly raised by the presence of subsidiary hospitals in the area. For example, if the returns from the subsidiary hospitals in North Canterbury are deleted, the admissions into Christchurch Hospital would represent about 38 per 1,000 ; a return not very much different from Auckland. I offer no criticism of these figures, as a much more minute examination would have to be made before reasonable conclusions could be drawn. However, even in their present form they offer a field for investigation.

27