- Email: [email protected]
The Role and Impact of Gender and Age on Children's Preferences for Pediatricians
The Role and Impact of Gender and Age on Children’s Preferences for Pediatricians Judith A. Turow, MD; Robert C. Sterling, PhD Objective.—To determine the physician gender preferences of children and the consequences of meeting/not meeting children’s preferences, both in their liking and feeling comfortable in talking with the pediatrician. Methods.—A convenience sample of 125 parent-child pairs completed surveys when coming for an outpatient visit to a university-sponsored, urban pediatric practice. Results.—Both adolescents and preadolescents (especially females) as young as 9 years of age expressed a gender preference for their physician. Meeting those preferences significantly affected how much children both liked and felt comfortable in talking with their physician. In young teen females, gender-preferred physicians are synonymous with gender-congruent physicians; yet although their preference for gender-congruent physicians increased in early adolescence, data indicated that their preference was often not met for several years to come. Conclusions.—Liking and comfort with the pediatrician are not only desirable, but may also influence the doctorpatient relationship and young people’s ability to develop health communication skills that they need as adults. These results, if validated, could also lead to a rethinking of parental dominance in the role of physician selection. KEY WORDS:
choice; gender preference; preference for physician
Ambulatory Pediatrics 2004;4:340 343
R
ecently, researchers have begun to isolate patient and physician factors that promote children’s involvement in, and satisfaction with, the health care delivery system.1–5 Kapphahn et al,3 for example, using national survey data, observed that 50% of adolescent female respondents reported a desire for a gender-congruent physician and that advancing adolescent age related to the likelihood of expressing such a preference. Other studies have shown patient/doctor gender congruence to contribute to a myriad of positive outcomes such as 1) improved communication, 2) patient satisfaction, and 3) decreased teen pregnancy rates.1,2,6 Similarly, Freed et al7 found a positive relationship between satisfaction with health care providers and intent to keep a follow-up appointment in a sample of 12–21-year-olds. Litt and Cuskey8 also observed that teens who were satisfied with their visit were more likely to return for care. Given the observed relationships between 1) patient age and desire for gender congruence, and 2) gender congruence and patient satisfaction, it seemed reasonable to examine the age at which children begin to profess such a wish. Two studies have demonstrated the ability of even young children to involve themselves in their health care. Lewis et al9 and Rudolph et al10 showed when parental
influences were minimized and children were allowed (required) to choose whether and how to access their health care, they accessed health care in a manner similar to adults as early as 5–8 years of age. Additionally, children who used the health care system on their own initiative were found to place greater emphasis on their own responsibilities for maintaining their health.9 The current pilot study was designed to examine the earliest age patients begin to express these gender preferences and whether meeting gender preferences influences children’s liking and comfort in speaking with their physician. Although some research has examined factors that influence adolescents’ preferences for their pediatrician, this topic remains understudied in a preteen population. In particular we asked: 1) Do patients of a specific gender choose physicians of a specific gender? 2) Does preference for physician gender change as the child grows, and if so, when? 3) Does meeting those preferences have an impact on how much the patient both likes and feels comfortable in talking with his/her physician? METHODS Procedure Paired surveys were distributed to a convenience sample of 415 nonconsecutive child-guardian dyads arriving for either a regularly scheduled well or sick outpatient visit at a university-based pediatric practice. Data for this study are drawn from the 125 parent-child pairs who returned completed surveys to the investigators. Children arriving without a parent/guardian were not offered participation in the study. Participants were informed that their responses were anonymous and that no identifying information should be placed on the surveys. Parents were also informed that
From the Department of Pediatrics (Dr Turow), AI duPont Children’s Hospital & Thomas Jefferson University, Wilmington, Del; and the Department of Psychiatry and Human Behavior (Dr Sterling), Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pa. Address correspondence to Judith A. Turow, MD, Assistant Professor of Pediatrics, Jefferson Medical College, Thomas Jefferson University, 833 East Chestnut St, Philadelphia, PA 19107. Received for publication July 18, 2003; accepted February 23, 2004. AMBULATORY PEDIATRICS Copyright q 2004 by Ambulatory Pediatric Association
340
Volume 4, Number 4 July–August 2004
AMBULATORY PEDIATRICS
Children’s Preferences for Pediatricians
341
Table 1. Relationship of Child Gender, Child Age, and Preference for Gender Congruent Physicians
Age
Male, n
% Males With MD Gender Preference
% Males with MD Gender Congruence
% Females with MD Gender Preference
% Females with MD Gender Congruence
Female, n
7–8 9–12 13–14 .15
20 21 11 10
30.0 33.5 18.2 20.0
50.0 47.6 27.3 70.0
35.3 61.9 92.3 77.8
70.6 61.9 69.2 88.9
17 21 13 9
x2 5 2.92, NS
x2 5 3.86, NS
x2 5 15.67, P , .05
x2 5 2.19, NS
although they were welcome to provide reading assistance as needed, they should try to refrain from actively influencing their children’s response(s). Participation was limited to families whose children were 7 years of age or older with prior experience with this practice. Parents arriving with multiple children were asked to provide answers with/for the one child being seen that day whose birthday was closest to the current visit. IRB approval from Thomas Jefferson University was obtained before study commencement. Survey Investigator-designed surveys consisting of a series of forced choice (n 5 4) and semantic differential (n 5 7) items were administered to parents and children. Items queried patients about their preferences regarding physician gender, and age and the degree to which they liked and felt comfortable with their doctor. Before survey administration, an expert panel of academic pediatricians at Jefferson/duPont reviewed the items for clarity, relevance, and appropriateness to child age. Items were previously pilot-tested on a sample of seventh- and eighth-grade respondents, and found to be clear and easily understood. Basic demographic information (eg, patient gender, age, and race, and family income) was obtained. Two items on the parent form assessed socioeconomic status: a self-report of familial income and an item that asked respondents to assess their economic well-being (from Kapphahn et al).3 Physicians Nine attending-level physicians (5 female) saw 118 survey respondents. Seven children were treated by one of 5 residents (all R3s, 4 female). The pediatric practice is staffed by full-time attending physicians, residents rotating through the practice, and continuity residents. The residents in this survey were all continuity residents, known to the patients and their parents ahead of time. The attending physicians were all known to the patients and their parents ahead of time. All of the physicians had been in practice at this site for a minimum of 3 years and all were white. Analysis First, using x2, we explored the relationship of child age and preference for a physician of a specific gender stratifying by child gender. Second, to determine whether children who had their preference met were more likely
to be satisfied with their physician, we contrasted their expressed preference for physician gender with the actual gender of their pediatrician, and examined the extent to which meeting those preferences affected the child’s liking of, and comfort, in talking with the physician. RESULTS One hundred ninety-three (46.5%) surveys were returned—125 sufficiently complete to allow inclusion. Numbers of cases available vary due to differential response rates to survey items. Patient Characteristics Fifty-one percent of the child respondents were male, 51% were African American, and 38.2% were white. Child respondents averaged 11.04 6 2.99 years of age. Sick visits accounted for 44.1% of patient visits, checkups for 50.8%. The remaining 5.1% were seen for immunizations or for form completion. Child Preferences Overall, 38% of patients (n 5 48) expressed a desire for a female physician, 8% (n 5 10) wished for a male, and 54% (n 5 67) indicated no preference. Female patients were significantly more likely to report a preference for their physician’s gender (63.4%) than were males (27%), x2 5 32.96, P , .001. Of the patients reporting a preference, almost all of the females (97.4%) wanted a gender-congruent (female) doctor as compared with only half (52.9%) of the males, x2 5 19.98, P , .001. Regression analyses indicated that these effects were independent of the effects of family income or race [F (3, 93) 5 .89, P . .05]. Table 1 shows the expression of gender preference to be age dependent in females, not males. The 62 male patients showed no significant relationship between their age and preference for physician gender, x2 5 2.92, P . .05. The 60 female patients, however, showed a preference emerging as early as 9 years of age, increasing markedly through 13–14 years, then decreasing slightly by age 15, x2 5 15.67, P , .05. Almost all of the females who expressed a preference wished for a gender-congruent pediatrician. We sought to examine whether these children were capable of both making their wishes known and influencing decisions regarding their provider as they matured and found that females’ likelihood of having secured a female doctor did not differ according to age. Given that female patients’
342
Turow & Sterling
AMBULATORY PEDIATRICS
Table 2. Associations of Meeting a Child’s Preference with Children’s Evaluations of Their Physician.
Like a lot Comfortable
No Preferences
Preferences Met
Preferences Not Met
P
87.3 95.4
88.6 91.0
61.5 76.9
.03 .04
desires for gender congruence increased in early adolescence (13–14-year-old group), it appeared as though these patients were not having this preference met. Child Evaluative Statements Of the 58 children who expressed a specific physician gender preference, 45 (36% of the overall sample) had a physician consistent with their gender preference; 13 (10% overall) did not. The remainder (n 5 67) reported no preference. As can be seen in Table 2, children whose gender preferences were not met were significantly less likely to report ‘‘liking their physician a lot,’’ (x2 5 6.97, P 5 .03), and ‘‘feeling comfortable in speaking with their physician,’’ (x2 5 6.29, P 5 .04) than those who either had no preference or had had their preferences met. DISCUSSION This study’s key findings are 1) children, especially females, as young as 9 years of age will express a preference for the gender of their pediatrician; 2) meeting those gender preferences can significantly affect children’s comfort with, and liking of, their pediatrician; and 3) in early adolescent females, where the desire for a gender-congruent physician is exceedingly high, what they get (the actual acquisition of the children’s preferred gender physician) appears to lag behind what they want (their expression of that preference) by several years. Given the relation between meeting preferences and feelings of comfort and liking, this discordance could seriously affect the doctor-patient relationship at a critical developmental moment. In the adult literature, embarrassment has been identified as a factor in wanting the same-gender physician.11,12 It may well be that similar cognitive processes are operating in maturing females. The lag between the initial expression of the desire for a gender-specific physician and the realization of this goal may speak to one of two factors: 1) the powerful connection that already exists between the parent and the children’s pediatricians, and 2) the lack of voice that the younger adolescent female has in choosing a physician. It is a shift of paradigm for the parents to attend to their children’s relationship to the pediatrician, over and above their own. It may be difficult and even unrealistic to expect parents to give up a physician whom they know and trust for a physician of their children’s choosing. It may take some time for parents to accustom themselves to the idea, plus perhaps some proof that the idea is a worthwhile one. The shift to a pediatrician of a children’s
choosing might be easier for parents who need to change physicians because of a move, or change of insurance. The generalizability of this study is hampered by a limited sample (both size and location—subject selection from a single site could possibly contribute to a selection bias). Reading level of the instrument (approximately 12 years) also implies that younger children may have needed to rely more on their parent(s) for guidance than we had wished. Thus, compelling as these results may be, we must acknowledge that potential parental influence on responses ultimately limits the degree to which we can be comfortable that answers reflected the child’s true feelings. In addition, the cross-sectional nature of this research does not allow for a discussion of a temporal causal connection—that is, initial desire for a particular gender of physician leads children to acquire a physician of that gender, and to therefore like and feel comfortable with that pediatrician. One can imagine a different interpretation of the associations. It is possible that children who are given pediatricians of a specific gender tend to like them and feel comfortable with them. As a result, when asked, they state that they would prefer a physician of that gender. That interpretation would not explain why a samegender physician is so much more important to female teens than to male teens. This also fails to explain why in early adolescent females the preference for a specific-gender physician often precedes the actual acquisition of such a desired physician, by numbers of years. Placing our findings in relation to previous work on physician preference leads us to suggest that the chain proceeds from preference to liking and comfort. Although potentially provocative, methodological limitations suggest the need for replication and cross-validation. It would also be interesting to observe whether meeting the preferences of children directly influenced outcomes such as overall client satisfaction and compliance. Certainly the validity of these results could be improved by having a neutral party administer the survey to child respondents. It, however, raises the possibility that allowing children to have a role in choosing a physician will yield benefits in child happiness and comfort with early health care relationships. Research by Lewis et al9 and Rudolph et al10 implies that children can be socialized into playing an active role in accessing and then maintaining their health care, in settings where parents had limited to no access to their child’s health care choices. It would be useful to know if self care along with medical compliance could be enhanced in more ordinary situations, if parents attended to their children’s basic physician preferences beginning as early as 9 years of age. This might engender children to become more engaged health care consumers throughout their lives. REFERENCES 1. Bernzweig J, Takayama J, Phibbs C, Pantell R. Gender differences in physician-patient communication, evidence from pediatric visits. Arch Pediatr Adolesc Med. 1997;151:586–591. 2. Boekeloo BO, Schamus LA, Cheng TL, Simmens SJ. Young
AMBULATORY PEDIATRICS
3.
4. 5.
6.
7.
adolescents’ comfort with discussion about sexual problems with their physician. Arch Pediatr Adolesc Med. 1996;150: 1146–1151. Kapphahn CJ, Wilson KM, Klein JD. Adolescent girls’ and boys’ preferences for provider gender and confidentiality in their health care. J Adolesc Health. 1999;25:131–142. Van Ness C. Male adolescents and physician sex preference. Arch Pediatr Adolesc Med. 2000;154:49–53. Stevens G, Shi, L. Effect of managed care on children’s relationships with their primary care physicians; differences by race. Arch Pediatr Adolesc Med. 2002;156:369–377. Hippsley-Cox J, Pringle M, Ebdon D, et al. Association between teenaged pregnancy rates and the age and sex of general practitioners: cross sectional survey in Trent 1994–7. Br Med J. 2000;320:842–845. Freed L, Ellen J, Irwin Jr. C, Millstein S. Determinants of adolescents’ satisfaction with their health care providers and their
Children’s Preferences for Pediatricians
8.
9.
10.
11.
12.
343
intentions to keep follow-up appointments. J Adolesc Health. 1998;22:475–479. Litt IF, Cuskey WR. Satisfaction with health care; a predictor of adolescents’ appointment keeping. J Adolesc Health Care. 1984;5:196–200. Lewis C, Lewis MA, Lorimer A, Palmer B. Child-initiated care: the use of school nursing services by children in an ‘‘adultfree’’ system. Pediatrics. 1977;60:499–507. Rudolf M, Tomanovich O, Greenberg J, et al. Gender differences in infirmary use at a residential day camp. J Dev Behav Pediatr. 1992;13:261–265. Heaton CJ, Marquez JT. Patient preferences for physician gender in the male genital/rectal exam. Fam Pract Res J. 1990;10: 105–115. Moettus A, Sklar D, Tandberg D. The effect of physician gender on women’s perceived pain and embarrassment during pelvic examination. Am J Emerg Med. 1999;17:635–637.
choice; gender preference; preference for physician
Ambulatory Pediatrics 2004;4:340 343
R
ecently, researchers have begun to isolate patient and physician factors that promote children’s involvement in, and satisfaction with, the health care delivery system.1–5 Kapphahn et al,3 for example, using national survey data, observed that 50% of adolescent female respondents reported a desire for a gender-congruent physician and that advancing adolescent age related to the likelihood of expressing such a preference. Other studies have shown patient/doctor gender congruence to contribute to a myriad of positive outcomes such as 1) improved communication, 2) patient satisfaction, and 3) decreased teen pregnancy rates.1,2,6 Similarly, Freed et al7 found a positive relationship between satisfaction with health care providers and intent to keep a follow-up appointment in a sample of 12–21-year-olds. Litt and Cuskey8 also observed that teens who were satisfied with their visit were more likely to return for care. Given the observed relationships between 1) patient age and desire for gender congruence, and 2) gender congruence and patient satisfaction, it seemed reasonable to examine the age at which children begin to profess such a wish. Two studies have demonstrated the ability of even young children to involve themselves in their health care. Lewis et al9 and Rudolph et al10 showed when parental
influences were minimized and children were allowed (required) to choose whether and how to access their health care, they accessed health care in a manner similar to adults as early as 5–8 years of age. Additionally, children who used the health care system on their own initiative were found to place greater emphasis on their own responsibilities for maintaining their health.9 The current pilot study was designed to examine the earliest age patients begin to express these gender preferences and whether meeting gender preferences influences children’s liking and comfort in speaking with their physician. Although some research has examined factors that influence adolescents’ preferences for their pediatrician, this topic remains understudied in a preteen population. In particular we asked: 1) Do patients of a specific gender choose physicians of a specific gender? 2) Does preference for physician gender change as the child grows, and if so, when? 3) Does meeting those preferences have an impact on how much the patient both likes and feels comfortable in talking with his/her physician? METHODS Procedure Paired surveys were distributed to a convenience sample of 415 nonconsecutive child-guardian dyads arriving for either a regularly scheduled well or sick outpatient visit at a university-based pediatric practice. Data for this study are drawn from the 125 parent-child pairs who returned completed surveys to the investigators. Children arriving without a parent/guardian were not offered participation in the study. Participants were informed that their responses were anonymous and that no identifying information should be placed on the surveys. Parents were also informed that
From the Department of Pediatrics (Dr Turow), AI duPont Children’s Hospital & Thomas Jefferson University, Wilmington, Del; and the Department of Psychiatry and Human Behavior (Dr Sterling), Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pa. Address correspondence to Judith A. Turow, MD, Assistant Professor of Pediatrics, Jefferson Medical College, Thomas Jefferson University, 833 East Chestnut St, Philadelphia, PA 19107. Received for publication July 18, 2003; accepted February 23, 2004. AMBULATORY PEDIATRICS Copyright q 2004 by Ambulatory Pediatric Association
340
Volume 4, Number 4 July–August 2004
AMBULATORY PEDIATRICS
Children’s Preferences for Pediatricians
341
Table 1. Relationship of Child Gender, Child Age, and Preference for Gender Congruent Physicians
Age
Male, n
% Males With MD Gender Preference
% Males with MD Gender Congruence
% Females with MD Gender Preference
% Females with MD Gender Congruence
Female, n
7–8 9–12 13–14 .15
20 21 11 10
30.0 33.5 18.2 20.0
50.0 47.6 27.3 70.0
35.3 61.9 92.3 77.8
70.6 61.9 69.2 88.9
17 21 13 9
x2 5 2.92, NS
x2 5 3.86, NS
x2 5 15.67, P , .05
x2 5 2.19, NS
although they were welcome to provide reading assistance as needed, they should try to refrain from actively influencing their children’s response(s). Participation was limited to families whose children were 7 years of age or older with prior experience with this practice. Parents arriving with multiple children were asked to provide answers with/for the one child being seen that day whose birthday was closest to the current visit. IRB approval from Thomas Jefferson University was obtained before study commencement. Survey Investigator-designed surveys consisting of a series of forced choice (n 5 4) and semantic differential (n 5 7) items were administered to parents and children. Items queried patients about their preferences regarding physician gender, and age and the degree to which they liked and felt comfortable with their doctor. Before survey administration, an expert panel of academic pediatricians at Jefferson/duPont reviewed the items for clarity, relevance, and appropriateness to child age. Items were previously pilot-tested on a sample of seventh- and eighth-grade respondents, and found to be clear and easily understood. Basic demographic information (eg, patient gender, age, and race, and family income) was obtained. Two items on the parent form assessed socioeconomic status: a self-report of familial income and an item that asked respondents to assess their economic well-being (from Kapphahn et al).3 Physicians Nine attending-level physicians (5 female) saw 118 survey respondents. Seven children were treated by one of 5 residents (all R3s, 4 female). The pediatric practice is staffed by full-time attending physicians, residents rotating through the practice, and continuity residents. The residents in this survey were all continuity residents, known to the patients and their parents ahead of time. The attending physicians were all known to the patients and their parents ahead of time. All of the physicians had been in practice at this site for a minimum of 3 years and all were white. Analysis First, using x2, we explored the relationship of child age and preference for a physician of a specific gender stratifying by child gender. Second, to determine whether children who had their preference met were more likely
to be satisfied with their physician, we contrasted their expressed preference for physician gender with the actual gender of their pediatrician, and examined the extent to which meeting those preferences affected the child’s liking of, and comfort, in talking with the physician. RESULTS One hundred ninety-three (46.5%) surveys were returned—125 sufficiently complete to allow inclusion. Numbers of cases available vary due to differential response rates to survey items. Patient Characteristics Fifty-one percent of the child respondents were male, 51% were African American, and 38.2% were white. Child respondents averaged 11.04 6 2.99 years of age. Sick visits accounted for 44.1% of patient visits, checkups for 50.8%. The remaining 5.1% were seen for immunizations or for form completion. Child Preferences Overall, 38% of patients (n 5 48) expressed a desire for a female physician, 8% (n 5 10) wished for a male, and 54% (n 5 67) indicated no preference. Female patients were significantly more likely to report a preference for their physician’s gender (63.4%) than were males (27%), x2 5 32.96, P , .001. Of the patients reporting a preference, almost all of the females (97.4%) wanted a gender-congruent (female) doctor as compared with only half (52.9%) of the males, x2 5 19.98, P , .001. Regression analyses indicated that these effects were independent of the effects of family income or race [F (3, 93) 5 .89, P . .05]. Table 1 shows the expression of gender preference to be age dependent in females, not males. The 62 male patients showed no significant relationship between their age and preference for physician gender, x2 5 2.92, P . .05. The 60 female patients, however, showed a preference emerging as early as 9 years of age, increasing markedly through 13–14 years, then decreasing slightly by age 15, x2 5 15.67, P , .05. Almost all of the females who expressed a preference wished for a gender-congruent pediatrician. We sought to examine whether these children were capable of both making their wishes known and influencing decisions regarding their provider as they matured and found that females’ likelihood of having secured a female doctor did not differ according to age. Given that female patients’
342
Turow & Sterling
AMBULATORY PEDIATRICS
Table 2. Associations of Meeting a Child’s Preference with Children’s Evaluations of Their Physician.
Like a lot Comfortable
No Preferences
Preferences Met
Preferences Not Met
P
87.3 95.4
88.6 91.0
61.5 76.9
.03 .04
desires for gender congruence increased in early adolescence (13–14-year-old group), it appeared as though these patients were not having this preference met. Child Evaluative Statements Of the 58 children who expressed a specific physician gender preference, 45 (36% of the overall sample) had a physician consistent with their gender preference; 13 (10% overall) did not. The remainder (n 5 67) reported no preference. As can be seen in Table 2, children whose gender preferences were not met were significantly less likely to report ‘‘liking their physician a lot,’’ (x2 5 6.97, P 5 .03), and ‘‘feeling comfortable in speaking with their physician,’’ (x2 5 6.29, P 5 .04) than those who either had no preference or had had their preferences met. DISCUSSION This study’s key findings are 1) children, especially females, as young as 9 years of age will express a preference for the gender of their pediatrician; 2) meeting those gender preferences can significantly affect children’s comfort with, and liking of, their pediatrician; and 3) in early adolescent females, where the desire for a gender-congruent physician is exceedingly high, what they get (the actual acquisition of the children’s preferred gender physician) appears to lag behind what they want (their expression of that preference) by several years. Given the relation between meeting preferences and feelings of comfort and liking, this discordance could seriously affect the doctor-patient relationship at a critical developmental moment. In the adult literature, embarrassment has been identified as a factor in wanting the same-gender physician.11,12 It may well be that similar cognitive processes are operating in maturing females. The lag between the initial expression of the desire for a gender-specific physician and the realization of this goal may speak to one of two factors: 1) the powerful connection that already exists between the parent and the children’s pediatricians, and 2) the lack of voice that the younger adolescent female has in choosing a physician. It is a shift of paradigm for the parents to attend to their children’s relationship to the pediatrician, over and above their own. It may be difficult and even unrealistic to expect parents to give up a physician whom they know and trust for a physician of their children’s choosing. It may take some time for parents to accustom themselves to the idea, plus perhaps some proof that the idea is a worthwhile one. The shift to a pediatrician of a children’s
choosing might be easier for parents who need to change physicians because of a move, or change of insurance. The generalizability of this study is hampered by a limited sample (both size and location—subject selection from a single site could possibly contribute to a selection bias). Reading level of the instrument (approximately 12 years) also implies that younger children may have needed to rely more on their parent(s) for guidance than we had wished. Thus, compelling as these results may be, we must acknowledge that potential parental influence on responses ultimately limits the degree to which we can be comfortable that answers reflected the child’s true feelings. In addition, the cross-sectional nature of this research does not allow for a discussion of a temporal causal connection—that is, initial desire for a particular gender of physician leads children to acquire a physician of that gender, and to therefore like and feel comfortable with that pediatrician. One can imagine a different interpretation of the associations. It is possible that children who are given pediatricians of a specific gender tend to like them and feel comfortable with them. As a result, when asked, they state that they would prefer a physician of that gender. That interpretation would not explain why a samegender physician is so much more important to female teens than to male teens. This also fails to explain why in early adolescent females the preference for a specific-gender physician often precedes the actual acquisition of such a desired physician, by numbers of years. Placing our findings in relation to previous work on physician preference leads us to suggest that the chain proceeds from preference to liking and comfort. Although potentially provocative, methodological limitations suggest the need for replication and cross-validation. It would also be interesting to observe whether meeting the preferences of children directly influenced outcomes such as overall client satisfaction and compliance. Certainly the validity of these results could be improved by having a neutral party administer the survey to child respondents. It, however, raises the possibility that allowing children to have a role in choosing a physician will yield benefits in child happiness and comfort with early health care relationships. Research by Lewis et al9 and Rudolph et al10 implies that children can be socialized into playing an active role in accessing and then maintaining their health care, in settings where parents had limited to no access to their child’s health care choices. It would be useful to know if self care along with medical compliance could be enhanced in more ordinary situations, if parents attended to their children’s basic physician preferences beginning as early as 9 years of age. This might engender children to become more engaged health care consumers throughout their lives. REFERENCES 1. Bernzweig J, Takayama J, Phibbs C, Pantell R. Gender differences in physician-patient communication, evidence from pediatric visits. Arch Pediatr Adolesc Med. 1997;151:586–591. 2. Boekeloo BO, Schamus LA, Cheng TL, Simmens SJ. Young
AMBULATORY PEDIATRICS
3.
4. 5.
6.
7.
adolescents’ comfort with discussion about sexual problems with their physician. Arch Pediatr Adolesc Med. 1996;150: 1146–1151. Kapphahn CJ, Wilson KM, Klein JD. Adolescent girls’ and boys’ preferences for provider gender and confidentiality in their health care. J Adolesc Health. 1999;25:131–142. Van Ness C. Male adolescents and physician sex preference. Arch Pediatr Adolesc Med. 2000;154:49–53. Stevens G, Shi, L. Effect of managed care on children’s relationships with their primary care physicians; differences by race. Arch Pediatr Adolesc Med. 2002;156:369–377. Hippsley-Cox J, Pringle M, Ebdon D, et al. Association between teenaged pregnancy rates and the age and sex of general practitioners: cross sectional survey in Trent 1994–7. Br Med J. 2000;320:842–845. Freed L, Ellen J, Irwin Jr. C, Millstein S. Determinants of adolescents’ satisfaction with their health care providers and their
Children’s Preferences for Pediatricians
8.
9.
10.
11.
12.
343
intentions to keep follow-up appointments. J Adolesc Health. 1998;22:475–479. Litt IF, Cuskey WR. Satisfaction with health care; a predictor of adolescents’ appointment keeping. J Adolesc Health Care. 1984;5:196–200. Lewis C, Lewis MA, Lorimer A, Palmer B. Child-initiated care: the use of school nursing services by children in an ‘‘adultfree’’ system. Pediatrics. 1977;60:499–507. Rudolf M, Tomanovich O, Greenberg J, et al. Gender differences in infirmary use at a residential day camp. J Dev Behav Pediatr. 1992;13:261–265. Heaton CJ, Marquez JT. Patient preferences for physician gender in the male genital/rectal exam. Fam Pract Res J. 1990;10: 105–115. Moettus A, Sklar D, Tandberg D. The effect of physician gender on women’s perceived pain and embarrassment during pelvic examination. Am J Emerg Med. 1999;17:635–637.