 |
 |
 |
|
|
The existence of geographic variations in the use of health care services has largely been documented in physical health (1–2). Fewer studies have examined variations in mental health services. Yet, geographic variations have been observed among states, provinces, and regions (3–6). Many factors can explain the variations in health services utilization. These factors include patient characteristics, health resources, and provider characteristics. Ultimately, geographic variations in health services utilization and resources are no problem if they reflect population needs. Studies among adults and children, however, have shown that mental health services utilization and resources do not reflect level of needs (4,7–10). Some studies of mental health services’ utilization and resources for children have focused exclusively on physician and hospital services, (that is, on what is available within the health care system) (4,9,11). Although many mental health problems among children are taken care of within the health care system, several are treated by nonphysicians (psychologists, social workers, and speech therapists), often within the school system (12). The availability of these nonphysician providers is rarely considered in studies that look at mental health resources. Mental health problems among children occur frequently in industrialized countries (13), but they are also common in the 2 largest provinces in Canada (14,15). For this reason, it is important that appropriate resources and services be available in due time to avoid suffering and further deterioration, because effective treatments and approaches do exist (16). There is some indication that current child mental health services in some regions do not correspond to the frequency of mental health problems and therefore not to the need for services, but this requires further confirmation (4). The current study attempted to verify this suspicion while avoiding some of the limits of previous studies. We set the following 2 objectives for this study: 1) to determine whether publicly funded mental health services and resources available in 4 large regions in the province of Quebec were distributed according to the mental health needs of children aged 6 to 14 years, as measured by a population survey, and 2) to assess whether the variations in mental health services and resources across the 4 regions had changed over a 5-year period. MethodsThis study was conducted in the province of Quebec (population 7.4 million). To provide reliable indicators for this study—especially of needs—we established 4 large regions out of 11 of the 18 Quebec health and social service regions. These 11 regions covered more than 90% of the province’s population. The population of these 4 regions ranges from 1.3 million to 2.1 million. For the other 7 health regions, data were unreliable or simply unavailable for 1 or more indicators. All indicators were calculated for each of the 4 large regions. For the purpose of the study, these regions were labelled as follows: metropolitan (Montreal-Centre and Laval), southeast (Montérégie and Estrie), central west (Outaouais, Lanaudière, Laurentides, and Mauricie-Bois-Francs), and central east (Québec, Saguenay-Lac St-Jean, and Chaudière-Appalaches). The correspondence between needs on one hand and resources and service use on the other hand was assessed with data from 1992–1993. Data on resources and service use were also obtained for 1997–1998 to compare with those for 1992–1993. Measures of Needs The study of morbidity led to assessing the most frequent Axis-I DSM-III-R disorders (anxiety disorders, depression, and externalizing disorders). Two interviewers visited the child’s home, where one questioned the child and the other, the parent. The Diagnostic Interview Schedule for Children (DISC), Version 2.25, was given to parents and adolescents aged 12 to 14 years (17). The Dominic Questionnaire was given to children aged 6 to 11 years (18). For externalizing disorders among children aged 6 to 11 years, the DISC was completed over the telephone with the teachers. At the end of the interview, the parents and the interviewers completed an adaptation scale, the Children’s Global Assessment Scale (CGAS) (19). The CGAS allows a rating from 0 to 100 for the functioning of the child. A score above 70 is considered to be in the normal range. The parents and adolescents also answered the following question on a 5-point Likert scale on the perception of a need for help: “To what extent do you think that (name of child) needs help, for whatever problem, from a doctor, a psychologist, a social worker, or from any other type of professional?” This question originates from the Service Utilization and Risk Factor Module of the Columbia University study (20). In the QCMHS, the answer to this question best predicted reported service use (12). Three indicators of morbidity were retained as measures of mental health needs and calculated for each of the 4 study regions: 1) the proportion of children with at least 1 mental health disorder, 2) the mean CGAS score, and 3) the proportion of children for whom there was a perception of a need for help. Parents are most likely to initiate the search for professional help for their child; therefore, to limit the amount of data presented, we used only the parents’ point of view. Measures of Professional Resources We assessed health professionals available in the community, based on data from their respective colleges. The most relevant professionals for whom data were available for both periods (1992–1993 and 1997–1998) were social workers, psychoeducators, pediatricians, psychiatrists, and family physicians. Data for psychologists were available only for the most recent period. Because it was impossible to distinguish professionals who were working with children from those working with adults or to determine the proportion of the time that they devoted to mental health services vs other services, rates of professionals (not FTE) per 100 000 population were calculated. Measures of Service Use More than 85% of family physicians or general practitioners across the province are paid on a fee-for-service (FFS) basis. Many psychiatrists, however, are paid either in part or in total on a sessional or annual salary basis, and this varies by region. Thus, FFS data underestimate the true amount of psychiatrist services. To correct for this underestimate, the number of FFS services that psychiatrists provided was increased in the following manner: if psychiatrists provided 1000 services on a FFS basis in a given region and if the remuneration for these services corresponded to 60% of the overall honoraria that the RAMQ paid to the psychiatrists in this region, we initially thought of dividing 1000 by 0.60 to obtain an estimate of the total number of services that psychiatrists possibly provided (that is, 1000/0.60 = 1667). Based on RAMQ information, however, it was estimated that approximately 45% of psychiatrists’ working time, paid on a sessional or salary basis, was devoted to clinical work. As a result, the number of FFS services was increased by 45% of the 667 services thought to have been provided by psychiatrists not paid on a FFS basis (0.45 x 667 = 300). Consequently, the corrected number of services by psychiatrists in that region was estimated at 1300 (1000 + 300). This corrected number of psychiatrist services was added to the number of services that family physicians provided. Overall rates of physician services for mental health problems per 1000 population were calculated by age group for both study periods. All hospital inpatient services in Quebec are recorded in a single database, maintained by the provincial Ministry of Health and Social Services. Using this database, rates of hospitalizations in acute care hospitals for specific ICD-9 mental disorders (that is, neurotic disorders, personality disorders, substance abuse, adjustment reaction, depression, and disturbance of conduct or emotions) were calculated for the 3 study age groups (age 6 to 8, age 9 to 11, and age 12 to 14 years) and for both study periods (April 1, 1992, to March 31, 1993, and April 1, 1997, to March 31, 1998). Because the information on mental health services comes from anonymous databases, it was impossible to determine how many of these services were repeat services provided to the same individual during the study period. Analysis Differences across regions were examined using the F-test for the CGAS and the Pearson chi-square test for the presence of mental health disorders and the perception of need for help. Contrary to measures of needs, resources and service utilization measures were based on complete population data, not sample data. Therefore, statistical differences among regions for resources and service utilization were not calculated. There was no need to test how close to the truth these differences were, because the measures were true population parameters, not sample statistics.
|
||||||