Family physician job satisfaction in different medical care organization models

Family Practice Vol. 17, No. 4 Oxford University Press 2000 Printed in Great Britain Family physician job satisfaction in different medical care organization models Carmen García-Peña a, Sandra Reyes-Frausto a, Isabel Reyes-Lagunes b and Onofre Muñoz-Hernández c García-Peña C, Reyes-Frausto S, Reyes-Lagunes I and Muñoz-Hernández O. Family physician job satisfaction in different medical care organization models. Family Practice 2000; 17: 309 313. Objectives. The aim of the present study was to estimate physician job satisfaction at the Mexican Institute of Social Security (IMSS), the Ministry of Health (SSA) and in the private sector, and to measure the association between these different family medical care organization models. Methods. A comparative cross-sectional design was used to investigate the job satisfaction of family physicians in private and institutional family medicine clinics. Satisfaction was measured with a previously constructed and validated instrument. The instrument measures the satisfaction in four areas: global satisfaction, institution where the physician works, the patients and themselves as physicians. Results. One hundred and seven IMSS physicians, 106 SSA physicians and 97 private physicians were selected randomly from a census according to the sample size. The sample was weighted. Fifty-one percent of IMSS and SSA physicians were dissatisfied, against 25% in the private sector, in the first three areas. Comparing the private model and the IMSS, differences were found (P 0.0001) in the area of global satisfaction [odds ratio (OR) = 2.47, 95% confidence interval (CI) 1.69 3.67], institution where the physician works (OR = 2.12, CI 1.45 3.13) and themselves as physicians (OR = 1.84, CI 1.28 2.65). When the private/ssa groups were compared, the differences were similar (P 0.0001). No differences were found in terms of the patients. When stratifying, the risks increased in females, in the group aged 31 40 years and in specialists in family medicine. Conclusions. The organization model is associated with dissatisfaction in all areas, except in the patients. Keywords. Family physician, job satisfaction, medical care models. Introduction The job satisfaction of the family physician is a critical factor for health systems because the primary care level is responsible for providing medical care to a greater proportion of the population than any other care level. Also, job satisfaction level could be related to the quality and efficiency of the care given. Satisfaction has been understood as the subject s perception of different areas, such as psychological and material rewards, relationship with patients, and social and intellectual work Received 7 June 1999; Revised 15 October 1999; Accepted 13 March 2000. a Department of Epidemiological and Health Services Research and c Medical Research Coordination, Mexican Institute of Social Security (IMSS) and b Psychology Faculty, National University of Mexico. Correspondence to CGP; Email: mcgp@cim.spin.comm.mx atmosphere. 1 Some other authors 2 have fragmented the concept even further to include the perception of autonomy, payment, number of working rules, relationship with patients and work peers, work load, prestige and status. The impact of bureaucracy and other aspects of models of medical care organizations on the physician s perception have been analysed by several authors. 3 6 Although some refer to the positive effect that a secure salary can have, the greater part agree that complex structures have negative effects on the physicians satisfaction, due basically to loss of autonomy. 7 Several authors have studied the relationship of satisfaction to some variables; however, the majority of the literature does not include family physicians, and the measuring instruments have been elaborated from only the investigators point of view. Additionally, the reports related to institutional environments, resulting in different characteristics from those in the health system 309

310 Family Practice an international journal in Mexico and Latin America. 8 Therefore, the objective of this investigation was to determine the association between the three organization models of medical care and the level of job satisfaction of the family physician, using a previously constructed and validated satisfaction measure instrument. 9 Methods A cross-sectional design with group comparison was used. The study population was made up of practising family physicians, not working in more than one of the investigated models, and who answered the inquiry in full. The sample frame was constructed by listing all the family physicians in Mexico City who work in family medicine clinics of the Mexican Instute of Social Security (IMSS), in the Ministry of Health (SSA) and in private family medicine clinics. The subjects were selected randomly. The sample size was calculated for a cross-sectional design with an α of 0.05 and β of 0.20, P o = 0.30 (satisfied subjects in IMSS), and P 1 = 0.54 (satisfied subjects in the private sector). Variables The independent variable was the medical care organization model, defined as the way to structure the work of health personnel. Three models were included: IMSS, SSA and the private system. The differences between the models can be seen in Table 1. The dependent variable was job satisfaction defined as the affective state related to the perception that the individual has in three areas: the institution where they work, their patients and themselves as physicians. General, economic and academic characteristics and work backgrounds were also studied. Job satisfaction measuring instrument The instrument used was constructed and validated previously in the population under study. 9 Eighty items measure the areas of global satisfaction, institution where the physician works, the patients and themselves as physicians. The methodology that we used has been documented previously. 9 Data analysis An intelligent capture system in dbase language was designed. The general variables among each of the institutions were compared using chi-square. The total of family physicians representing each group was considered. The means obtained from the validation in the satisfaction instrument were taken as reference, and the proportion of satisfaction, by areas, and for each group, was obtained. A bivariate analysis with chi-square was carried out, obtaining odds ratios (ORs) and 95% confidence intervals (CIs). A stratified analysis using the general variables with Mantel and Haenzel was done. Results Population description A census was taken of 1447 family physicians in 42 IMSS family medicine clinics, 504 physicians in the SSA, and 150 physicians in private family medicine clinics. A total of 107 IMSS physicians (1:13.5 of total), 106 from the SSA (1:4.7) and 97 private physicians (1:1.6) were included. The rate of non-response was 10% in all groups. More than half of the physicians in all groups were males; among private physicians, the proportion reached 75.3% (P 0.03). Nearly 50% of private and SSA physicians were between 36 and 40 years of age; in the case of IMSS physicians, the majority were between 36 and 45 years of age. The mean age for IMSS physicians was 42 years, for SSA it was 40 years and for private physicians, 39 years (P 0.03). Almost all physicians were married, the highest percentage (84.5%) being found for private physicians (P 0.03). Fifty-three percent (n = 57) of the IMSS physicians, 49% (n = 48) of private physicians and 23% (n = 25) of the SSA physicians had specialized in family medicine (P 0.001) (Table 2). The proportions of satisfied physicians were higher in the private physicians group in all areas except the patients, and varied between 63 and 73%. The lowest proportion (45%) of satisfied physicians was that of the SSA in the themselves as physicians area, but in the patients, this model had the highest proportion of satisfied physicians (67%) (Table 3). TABLE 1 General characteristics of the care models Model Characteristics IMSS SSA Private Bureaucracy Relationship of physician organization Employee Employee Independent Accessibility Funding/budget Government, industry, worker Government Banking system Relationship of patient organization Insured Open Insured

Family physician job satisfaction 311 TABLE 2 General characteristics of the sample Variables IMSS SSA Private P-value n = 107 n = 106 n = 97 % % % Gender Male 61.7 57.5 75.3 0.02 Female 38.3 42.5 24.7 Age 20 26 1.9 11.3 3.1 0.03 31 35 16.8 12.3 15.5 36 40 27.1 41.5 47.4 41 45 27.1 17.0 22.7 46 50 12.1 9.4 8.3 51 55 8.4 1.9 1.0 56 and over 6.4 6.6 2.0 Average 42.28 40.15 39.8 Marital status Single 13.1 22.6 11.3 Married 83.1 69.8 86.6 0.01 Widowed 1.9 1.9 Family medicine speciality Yes 53.3 23.6 49.5 No 46.7 76.4 50.0 0.000 TABLE 3 Proportion of satisfaction Dimension IMSS SSA Private n = 107 n = 106 n = 97 % % % Global satisfaction Satisfied 52 54 73 Dissatisfied 48 46 27 Health institute Satisfied 55 50 72 Dissatisfied 45 50 28 The patients Satisfied 57 67 63 Dissatisfied 43 33 37 Themselves as physicians Satisfied 51 45 65 Dissatisfied 49 55 35 Bivariate analysis Comparing the SSA and private groups in the area of global satisfaction, the probability of no satisfied physicians was 2.33 more in the SSA than in the private group (95% CI 1.55 3.55); for dissatisfaction with the institution where the physician works, the SSA showed a probability of 2.60 (95% CI 1.73 3.96). For themselves as physicians, the probability of dissatisfaction was 2.26 (95% CI 1.53 3.35). The probabilities obtained for these areas were similar to those obtained when comparing private and IMSS physicians. Dissatisfaction probabilities were not obtained for the patients for the SSA (OR = 0.83, 95% CI 0.56 1.23). Comparing the IMSS group with SSA, there was a statistically significant dissatisfaction probability of 1.22 in the institution where the physician works (95% CI 1.00 1.50) and of 1.23 for themselves as physicians (95% CI 1.00 1.51) (Table 4). Stratified analysis When comparing the private IMSS groups, being female was maintained as a constant dissatisfaction risk for all areas, except for the patients, with probabilities that varied between 3.13 and 5.07, all with statistically significant intervals. Similar dissatisfaction probabilities for the same areas were obtained in the age groups of between 31 and 40 years. Being a specialist in family medicine increased the probability of dissatisfaction to 3.19 in the global satisfaction area (95% CI 1.76 6.04, P = 0.000). In the private SSA groups, the probability of dissatisfaction was increased in the global satisfaction and institution where the physician works areas from 2.51 to 4.25 due to being a specialist in family medicine, of female gender and/or of an age between 36 and 40 years (P 0.000). Being aged between 31 and 35 years and between 31 and 40 also increased the dissatisfaction probability for institution where the physician works and themselves as physicians to 4.08 and 7.46, respectively (P 0.000). Unlike the private IMSS groups, the fact of not being a specialist in family medicine increased the dissatisfaction probability for themselves as physicians to 2.97 (95% CI 1.74 5.17, P = 0.001) in the private SSA groups. When stratifying, the comparison between IMSS and SSA groups did not give statistically significant data (Table 5). Discussion In 1969, Engels 4 studied the impact that bureaucracy had on the job satisfaction of physicians. He found that complex structures transformed the nature of the physicians task, and focused the physicians attention towards the institutions and not towards the patients. The starting point of this study is based on this report written 25 years ago. There are several aspects that make comparison with our results difficult. There are very few references from Mexico, and none compare different organization models; on the other hand, the methodology followed for the construction and validation of the measuring instrument

312 Family Practice an international journal TABLE 4 Bivariate analysis (weighted) according to dimension by model of care Dimension Private IMSS Private SSA IMSS SSA OR 95% CI OR 95% CI OR 95% CI Global satisfaction 2.47* 1.69 3.67 2.33* 1.55 3.55 0.94 0.76 1.15 Health institute 2.12* 1.45 3.13 2.60* 1.73 3.96 1.22* 1.00 1.50 The patients 1.27 0.90 1.82 0.83 0.56 1.23 0.65 0.37 1.14 Themselves as physicians 1.84* 1.28 2.65 2.26* 1.53 3.35 1.23** 1.00 1.51 *P 0.0001; **P = 0.04. TABLE 5 Stratified analysis according to dimension by model of care Dimension Variable IMSS Private SSA Private OR 95% CI OR 95% CI Global satisfaction Family medicine specialist 3.19* 1.78 6.04 4.15* 2.06 8.59 31 35 years 4.18* 1.48 13.45 36 40 years 3.96* 2.15 7.62 3.38* 1.77 6.70 Female 3.24* 1.42 8.33 2.51** 1.06 6.63 Health institute Family medicine specialist 1.99* 1.18 3.44 2.80* 1.47 5.39 31 35 years 5.65* 1.92 19.99 5.70* 1.71 21.84 36 40 years 2.76* 1.57 5.00 4.08* 2.22 7.65 Female 5.07** 2.04 15.04 4.25* 1.47 5.39 Themselves as physicians 31 35 years 12.67* 3.54 68.12 7.46* 1.87 42.47 36 40 years 1.96* 1.12 3.52 4.24* 2.31 7.89 Female 3.13* 1.51 6.67 2.97* 1.74 5.17 *P 0.006; **P 0.02. has not been used in the investigation area of health services. Private physicians were the most satisfied group in three of the four areas. The groups of physicians from IMSS and SSA had a lower and very similar level of satisfaction in the areas of global satisfaction and institution where the physician works. These findings coincide with what has been reported in the literature. 10,11 It is known that low autonomy, excessive norms and stress are aspects related to job dissatisfaction and, therefore, it is to be expected that not only the IMSS but also the SSA have a lower proportion of satisfied physicians. In the area of the patients, the proportion of satisfied physicians was somewhat higher in the SSA than in the private and IMSS groups, findings that were confirmed in the bivariate and stratified analyses. No differences existed in the perception that physicians have of their patients in any of the three groups, i.e. the physician patient relationship is an important source of satisfaction. 12,13 In the themselves as physicians area, the highest proportion of satisfied physicians was found in the private group; it was lower in the IMSS, and even lower in the SSA, which is confirmed in the bivariate analysis. This area is related to the perception that the physicians have of themselves regarding knowledge and status levels. We think that the institutional characteristics have influenced the loss of self-esteem, as the physicians in both the IMSS and SSA must comply with a series of rules in order to render care, which has restricted the freedom of autonomy and self-responsibility. The relationship between the level of satisfaction and the models does not decrease in any of the areas when stratifying with the intervening variables. In both the SSA and IMSS, being female increased the probability of dissatisfaction in most of the areas. The findings in the literature 14,15 are contradictory, but we think that a greater difficulty in obtaining professional recognition for women is an important factor in this finding. The dissatisfaction probability is also increased in the young

physician group. It is possible that the young physicians have greater demands and, as age advances, the possibility to adapt increases. 16 Being a specialist in family medicine produced a greater dissatisfaction probability in the IMSS and SSA. There are no reports in the literature on this subject, but we suppose that specialists have greater improvement perspectives and, on entering a model that does not meet these expectations, they feel more dissatisfied. For the area themselves as physicians, the dissatisfaction probability is almost twice as high, as it is more difficult to become a specialist in the SSA model, and this fact can be seen as a self-fulfilment possibility. The dissatisfaction probabilities obtained for the IMSS and SSA models speak favourably to the fact that bureaucratic organizations frequently are in discordance with the professionals who work there. 17 It is probable that the situational aspects, 18 i.e. the extrinsic aspects, on the individual most frequently are related to job dissatisfaction, as none of the individual characteristics resulted in the model. In this regard, aspects such as personality, mental health, etc. have little relationship to job satisfaction. However, we know that future investigations must delve into the study of intrinsic factors and their relationship to job satisfaction, as it will also be necessary to explore the satisfaction related to the quality of the care given. We consider that the future challenge dwells in diminishing bureaucracy in health institutes, and in increasing autonomy, protecting the social security and equality principles in such a manner that the satisfaction of the family physician can be increased. A full copy of the questionnaire used in this study can be obtained from the corresponding author. Acknowledgements The project was supported by the National Council of Science and Technology (CONACYT). Family physician job satisfaction 313 References 1 Kravitz R, Linn L, Shapiro M. Physician satisfaction under the Ontario Health Insurance Plan. Med Care 1990; 28: 502 512. 2 Stamps P, Piedmont E et al. Measurement of work satisfaction among health professionals. Med Care 1978; 16: 337 343. 3 Staw B, Ross J. Stability in the midst of change: a dispositional approach to job attitudes. J Appl Psychol 1985; 70: 469 480. 4 Engel GV. The effect of bureaucracy on the professional autonomy of the physician. J Health Soc B 1969; 10: 30 40. 5 Mechanic D. General medical practice: some comparison between the work of primary care physicians in the United States and England and Wales. Med Care 1972; 10: 402. 6 Breslau N, Novack A. Work setting and job satisfaction. Med Care 1978; 16: 850 862. 7 Cooper C, Rout V. Mental health, job satisfaction and job stress among general practitioners. Br Med J 1989; 298: 366 370. 8 Aguirre-Gas H. Evaluación de la calidad de la atención médica, expectativas de los pacientes y de los trabajadores en las unidades médicas. Salud Pública Mex 1990; 32: 170 180. 9 Garcia-Peña C, Reyes-Lagunes I, Reyes-Frustro S, Villa-Contreras S, Libreros-Bango V, Muñoz-Hernández O. Development and validation of an inventory for measuring job satisfaction among family physicians. Psychol Rep 1996; 79: 291 301. 10 Lichtenstein R. Measuring the job satisfaction of physicians in organized settings. Med Care 1984; 22: 56 68. 11 Burns L, Anderson R, Shortell S. The effect of hospital control strategies on physician satisfaction and physician hospital conflict. Hosp Ser Res 1990; 25: 527 560. 12 Shulz R, Girard C, Scheckler W. Physician satisfaction in a managed care environment. J Fam Pract 1992; 34: 298 304. 13 Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Fam Pract 1993; 37: 257 263. 14 Johnson N, Hasler J, Mant D, Randall T, Jones L, Yudkin P. General practice careers: changing experience of men and women vocational trainees between 1974 and 1989. Br J Gen Pract 1993; 43: 141 145. 15 Lewis JM, Nace EP, Barnhart FD, Carson DI, Howard BL. The lives of female physicians. Tex Med 1994; 90: 56 61. 16 Groenwegen P, Hutten J. Workload and job satisfaction among general practitioners: a review of the literature. Soc Sci Med 1991; 32: 1111 1119. 17 Lichtenstein R. The job satisfaction and retention of physicians in organized settings. A literature review. Med Care Rev 1984; 41: 139 179. 18 Gruenberg B. The happy worker: an analysis of educational and occupational differences in determinants of job satisfaction. Am J Sociol 1980; 86: 247 271.