Measuring the burden of preventable diabetic hospitalisations in the Mexican Institute of Social Security (IMSS)

David G. Lugo-Palacios, John Cairns, Cynthia Masetto

The prevalence of diabetes among adults in Mexico has increased markedly from 6.7 % in 1994 to 14.7 % in 2015. Although the main diabetic complications can be prevented or delayed with timely and effective primary care, a high percentage of diabetic patients have developed them imposing an important preventable burden on Mexican society and on the health system. This paper estimates the financial and health burden caused by potentially preventable hospitalisations due to diabetic complications in hospitals operated by the largest social security institution in Latin America, the Mexican Institute of Social Security (IMSS), in the period 2007-2014. Hospitalisations in IMSS hospitals whose main cause was a diabetic complication were identified. The financial burden was estimated using IMSS diagnostic-related groups. To estimate the health burden, DALYs were computed under the assumption that patients would not have experienced complications if they had received timely and effective primary care. A total of 322,977 hospitalisations due to five diabetic complications were identified during the period studied, of which hospitalisations due to kidney failure and diabetic foot represent 78 %. The financial burden increased by 8.4 % in real terms between 2007 and 2014. However, when measured as cost per IMSS affiliate, it decreased by 11.3 %. The health burden had an overall decrease of 13.6 % and the associated DALYs in 2014 reached 103,688. Resources used for the hospital treatment of diabetic complications are then not available for other health care interventions. In order to prevent these hospitalisations more resources might need to be invested in primary care; the first step could be to consider the financial burden of these hospitalisations as a potential target for switching resources from hospital care to primary care services. However, more evidence of the effectiveness of different primary care interventions is needed to know how much of the burden could be prevented by better primary care.