The Hospital Morbidity Survey (HMS) performed in hospitals is a sample study that began in 1951 and has been carried out annually since then.
Its main objective is to ascertain the structure and development of hospital morbidity, that is to say, all persons admitted to a hospital centre for treatment as in-patients.
In particular, the Survey aims to measure morbidity according to the main or discharge diagnosis, the average hospital stay by type of diagnosis, the scope of influence or degree of attraction of hospitals and to act as a starting point for in-depth studies on certain diseases.
It was originally obligatory only for public hospitals to take part in the Survey, with the participation of private centres being voluntary, but the Survey was reviewed in 1977 and became a national Survey, including all Spanish public and private hospitals.
The fulfilment of the Survey's objectives was guaranteed with the introduction of the obligatory use of basic health documents in 1977: the Patient Register and the Patient File, which granted the systematic recording of all patients who were admitted to and discharged from hospital.
In the last few years, a new hospital information system has been developed for patient discharge, this being called the Basic Minimum Set of Data (BMSD), which responds to the needs of multiple users (management, planning, epidemiological research, statistical purposes, etc.) and which includes a number of administrative and clinical data collected when the patient is discharged. This system makes homogenous information between different hospitals available and covers the Survey objectives to the same extent that the Patient Register did in the past.
The classification variables of the Survey consist of the main diagnosis and the demographic and clinical circumstances of the patient (sex, age, date of admission and discharge, province of residence, type of admission, reason for discharge). This allows us to understand morbidity in hospitals in a certain reference year and makes information available on a state level on the availability and use of hospital resources. From 2011 on, information broken down has been included to the publication according to the Health dependency, showing information about public and private hospitals separately
Currently, the collection of necessary information in hospitals for the Survey is mixed, as the information comes from the two aforementioned base documents: the BMHDS, for both public and private hospitals, and the Record Book, mainly for those centres that lack computerised monitoring of patients.
Type of research
From the start, this statistical operation has been carried out by sampling and the selection is undertaken in two stages: in the first stage, a sample of hospitals is obtained and in the second stage, the discharged patients are selected. Currently, the hospital sample represents 95.4% of the total hospital centres and the number of patients admitted obtained from this sample exceeds 4.5 million people, which represents 99.2% of total annual hospital discharges in Spain.
In the last few years, the hospital information has been increasingly computerised, including the Register and the BMSD, which has facilitated the electronic collection of the data.
In terms of the hospital sample, at least one hospital from each of the objectives (general, surgical, children, etc.) is represented in each province, given the existing direct relationship between the main diagnosis and the type of care provided by the hospital centre.
As of HMS 2005, it is included in the publication of the shortlist of diagnostic groups for international data collection, agreed by Eurstat, OECD and WHO, for the dissemination of Hospital Morbidity data, which allows for the homogenous comparison of figures and diagnoses among countries and includes, for individual diagnoses or groups of them, 149 items with a four-digit level of development according to International Classification of Diseases ICD-9-CM (more information in Eurostat website http://www.ec.europa.eu/eurostat). From that year on, there is also microdata with a breakdown level of four digits on the main diagnosis.