Journal of Adolescent Health
Volume 42, Issue 2 , Pages 146-153, February 2008

The Parental Monitoring of Diabetes Care Scale: Development, Reliability and Validity of a Scale to Evaluate Parental Supervision of Adolescent Illness Management

  • Deborah A. Ellis, Ph.D.

      Affiliations

    • Carman and Ann Adams Department of Pediatrics, Wayne State University, Detroit, Michigan
    • Corresponding Author InformationAddress correspondence to: Deborah A. Ellis, Ph.D., Pediatric Prevention Research Center University Health Center 6-D, 4201 St. Antoine, Detroit, MI 48201.
    • ,
  • Thomas N. Templin, Ph.D.

      Affiliations

    • Center for Health Research, College of Nursing, Wayne State University, Detroit, Michigan
    • ,
  • Cheryl-Lynn Podolski, Ph.D.

      Affiliations

    • Department of Psychiatry and Behavioral Neurosciences, Wayne State University, Detroit, Michigan
    • ,
  • Maureen A. Frey, Ph.D.

      Affiliations

    • Carman and Ann Adams Department of Pediatrics, Wayne State University, Detroit, Michigan
    • ,
  • Sylvie Naar-King, Ph.D.

      Affiliations

    • Carman and Ann Adams Department of Pediatrics, Wayne State University, Detroit, Michigan
    • ,
  • Kathleen Moltz, M.D.

      Affiliations

    • Carman and Ann Adams Department of Pediatrics, Wayne State University, Detroit, Michigan

Received 8 March 2007; accepted 17 July 2007. published online 13 December 2007.

Article Outline

Abstract 

Purpose

Monitoring of adolescents’ behavior and whereabouts has been repeatedly identified as an important predictor of adolescent behavioral outcomes. However, to date, measures of parental supervision and monitoring are lacking in the chronic illness literature. The present study describes development and initial evaluation of a measure of parental monitoring of the illness management of adolescents with diabetes: the Parental Monitoring of Diabetes Care scale (PMDC).

Methods

Ninety-nine parents of 12–18-year-old children with type 1 diabetes completed the PMDC. Measures of illness management and metabolic control were also obtained.

Results

The PMDC demonstrated good internal consistency (alpha coefficient = .81) and test–rest reliability (ICC = .80). Supporting the instrument’s construct validity, confirmatory factor analysis indicated that a five subdomain structure had an acceptable fit to the data, [χ2 (181.65)/df (126) = 1.44, Bollen-Stine χ2 = 165.03, p = .32, comparative fit index (CFI) = .91, and root-mean-square error of approximation = .07]. In structural equation models, parental monitoring as assessed by the PMDC had a significant direct effect on adolescent diabetes management, accounting for 38% of the variance. Parental monitoring also had a significant indirect effect on metabolic control.

Conclusions

The PMDC represents an important first step in the development of measures of parental monitoring for use with adolescents with chronic medical conditions.

Keywords: Parental monitoring, Adolescents, Diabetes, Illness management

 

Parental monitoring refers to those aspects of parenting behavior that involve information seeking about the youth’s daily activities as well as direct supervision and oversight of those activities. Monitoring of adolescents’ behavior and whereabouts has been repeatedly identified as an important predictor of adolescent behavioral outcomes. Poor monitoring by parents has been found to predict early sexual initiation [1], school failure [2], use of alcohol and drugs [3], [4], [5], and involvement in antisocial activities [6], [7]. As a result, interventions to reduce risky behaviors in youth often attempt to increase parental oversight and supervision of adolescent activities [8], [9].

Behaviors that have conceptual similarity to parental monitoring and supervision have also been linked to illness management and physical health outcomes among adolescents with chronic medical illness. In particular, higher levels of parental involvement in the medical regimen have been shown to be related to better illness management practices in chronically ill youth [10], [11], [12], [13], as well as to better metabolic control in youth with diabetes [14], [15] and to lower functional morbidity in adolescents with asthma [16].

Despite the fact that parenting behaviors similar to parental monitoring appear to have important implications for the health outcomes of chronically ill adolescents, and despite the ubiquity of measures of parental monitoring in the general adolescent development literature [17], [18], [19], to date, there has been a dearth of measures of parental monitoring in the chronic illness literature. Existing measures of parental involvement in the adolescent’s medical regimen have typically evaluated whether the adolescent or the parent holds primary responsibility for completion of a given aspect of illness management [10], [14], [20]. Hence, such assessment tools are not designed to determine whether adolescents who hold partial or primary responsibility for their own medical care are monitored by their parents.

The purpose of the present report was to describe the development and initial evaluation of a parent-report measure of parental monitoring of the illness management of adolescents with diabetes: the Parental Monitoring of Diabetes Care scale (PMDC). We chose to develop an illness-specific measure rather than one that could be used across chronic conditions because it seemed plausible that increasing the specificity of questions related to parental monitoring of the regimen by asking about illness-specific tasks would increase the measure’s validity. Diabetes was chosen as the illness for instrument development because it is one of the most common childhood chronic illnesses and because the prior literature on youth with diabetes suggests that parenting behaviors in this population are related to youths’ illness management behavior and health outcomes [21].

Back to Article Outline

Methods 

Participants 

Participants were recruited from a university-affiliated pediatric diabetes clinic located in a tertiary care hospital facility in a large Midwestern metropolitan area. This convenience sample was recruited by approaching potential participants in person at the time of a regularly scheduled clinic visit. The research was approved by the Human Investigation Committee of the university affiliated with the hospital where the adolescents were seen for medical care. All participants provided informed consent and assent to participate.

To be eligible for the study, participants had to be the primary caregiver of an adolescent who was (1) between 12 and 18 years of age, (2) diagnosed for at least a year with type 1 diabetes (3) with no known developmental delay or other chronic medical conditions, and (4) residing in the participant’s home. Participants also had to be English speaking. Of the 128 potential participants who were approached, 103, or 80% of those eligible, agreed to participate. The most frequent reason for nonparticipation was the extra time required during the clinic appointment to complete the research measures. Four of the consented participants were later found to be ineligible due to youth cognitive limitations or physical limitations (e.g., one youth had cerebral palsy and did not administer his own insulin) that became apparent when data collection was initiated and were subsequently excluded. The final sample consisted of 99 participants. Both adolescent and parents provided data for the larger study, including data on parental monitoring; however, the present study focuses only on the psychometric properties of the parent-report version of the PMDC.

Demographic characteristics of the participants are shown in Table 1. Seventy-eight percent of parents were female. Forty-seven percent of parents were white, 36% were African American, and the rest were of other race/ethnicity. Mean family income was $43,625, and the majority of adolescents resided in two-parent homes (65%). Overall, the demographics of the sample were representative of the diverse, urban population served by the clinic where subjects were recruited.

Table 1. Demographic characteristics of study participants (n = 99)
%M (SD)
Youth age 14.8(1.7)
Parent Age 43.2(6.9)
Annual family income (dollars) $43,625(26,024)
Youth gender
Male52
Female48
Caregiver gender
Male22
Female78
Number of parents in home
Two parents65
Single parent35
Family ethnicity
Caucasian47
African American36
Other/missing17
Duration of diabetes in years 5.7(4.2)
HbA1c 9.1(2.3)
Insulin regimen
2–3 injections/day24
4 or more injections/day52
Insulin pump24

Procedure 

Development of PMDC scale 

Items were generated by a two-steps process. First, three experienced diabetes researchers (two doctoral level pediatric psychologists and one doctoral level nurse) developed a pool of items based on items from existing instruments and expert consensus. Items were generated in five subdomains. The first three subdomains captured the range of diabetes management needs and tasks that a parent might be expected to monitor or supervise. These were supervision of (a) the availability of medical supplies/devices, (b) checking blood glucose, and (c) diet. These subdomains were intended to capture parental monitoring across different areas of the regimen; however, exercise items were not included because although exercise can help lower blood glucose, specific prescriptions or recommendations for daily exercise are not typically provided to youth with type 1 diabetes. Items typifying the three subdomains included, “When your child runs out of test strips, how quickly do you know?” “How often do you review the readings in the blood glucose meter with your child” and “How often do you eat meals with your child?” A fourth domain included items to evaluate parental awareness of episodes of nonadherence. A sample item in this domain was “When your child skips insulin, how quickly do you know?” Items assessing monitoring of youth insulin administration behaviors were included in the nonadherence subdomain rather than in the first three subdomains due to widely varying ways that youth calculate and administer insulin depending upon their regimen (e.g., nonintensive injection regimen, balas–bolus injection regimen, insulin pump). Hence, an item evaluating parental knowledge of missed insulin doses was applicable to all parents regardless of the youth’s insulin regimen. Finally, items were generated to evaluate whether or not the parent supervised the child by direct oversight. A sample item in this domain was “How often are you present in the room when your child takes insulin?” Next, the item pool was reviewed by a pediatric endocrinologist and two certified diabetes nurse educators who generated additional items in the five domains. Eighteen items were retained for administration in the present study based on expert consensus across the evaluators that they had face validity. Parents were asked to think about their monitoring behavior during the past month when answering the questions. Item response was on a five point Likert scale (e.g., “More than once a day,” “Once a day,” “Several times a week,” “Once a week,” and “ Less than once a week”).

Data collection 

All measures were collected by a trained research assistant. Participants completed measures during a scheduled visit to the diabetes clinic. A subset of participants (n = 25) also completed the monitoring measure a second time 2 weeks after the initial completion of the questionnaire to evaluate test–retest reliability. The questionnaire was mailed to parents who were asked to return it in a self-addressed, stamped envelope; however, if it was not returned within a week of the due date, an attempt was made to complete the questionnaires by telephone. Six participants (24%) completed the questionnaire over the phone. Participants were provided with a $5 gift certificate for each portion of the study in which they participated.

Measures 

Illness management 

The study used a multi-informant, multimethod approach to measuring illness management. Parent and adolescent report of illness management as well as an objective measure of illness management were obtained. The Diabetes Management Scale (DMS; [22]) is a 20-item questionnaire designed to measure a broad range of diabetes management behaviors, such as insulin management, dietary management, blood glucose monitoring, and symptom response. Respondents are asked “What percent of the time do you (take your insulin)” and answer on a 0%–100% scale. Items are summed to obtain a total score reflecting overall management behavior. The instrument has demonstrated adequate reliability and validity [23], [24]. For the present study, adolescents rated their own diabetes management and parents completed a parallel form on the adolescent’s diabetes management.

Like most questionnaire measures of diabetes management, the DMS was developed prior to the widespread use of basal–bolus insulin regimens. Therefore, a subset of dietary items in the DMS, such those that ask about adherence to a prescribed meal plan, were not appropriate for adolescents on basal–bolus regimens and these items were not used in the present study. The alpha coefficient for the version of the DMS used in the present study was .72 for parent report and .70 for adolescent report.

Illness management was also measured by frequency of blood glucose (BG) checking. Data were obtained directly from the adolescent’s blood glucose meter to obtain the most objective information possible. The frequency of checking during the 14-day period immediately preceding data collection was recorded and an average daily checking frequency was subsequently calculated.

Metabolic control 

Metabolic control was calculated using hemoglobin A1c (HbA1c), a retrospective measure of average blood glucose during the past 2 to 3 months. Values were obtained during the medical appointment in the diabetes clinic with a DCA 2000 system (Bayer, Elkhart, IN) that uses an immunoglobulin-agglutination methodology.

Statistical analyses 

Reliability of the PMDC was determined by Cronbach’s alpha. In addition, test–retest reliability was evaluated using a subset (n = 25) of the cases by intraclass correlation (ICC). Construct validity was examined by conducting a confirmatory factor analysis (CFA) to test the hypothesis that the PMDC measured a unitary underlying parental monitoring construct that was comprised of five monitoring subdomains (supervision of the availability of medical supplies/devices, monitoring of blood glucose (BG) checking, oversight of diet, monitoring of nonadherence, and direct oversight of diabetes management behaviors). To accomplish this, a second-order CFA was performed on the variance–covariance matrix of the 18 items of the PMDC. Because the items exhibited significant skewness, estimation was performed using the parametric bootstrap procedure in AMOS 7.0 [25]. The bootstrap procedure is expected to perform better in small samples with skewed data than asymptotic methods [26], [27].

The concurrent validity of the PMDC was examined using structural equation modeling (SEM) where parental monitoring as evaluated by the PMDC was used to predict diabetes management and metabolic control. To reduce the number of parameters for this model, a simplified measurement model was used that consisted of one composite indicator for each subdomain—the subscale score. This approach is similar to the method of aggregating items to form parcels [28], although parcels are usually constructed to represent individual items rather than subdomain factors. In this model, PMDC was conceptualized as the sole exogenous variable, while diabetes management and metabolic control were endogenous. The three measures of illness management, parent report, adolescent report, and BG checking frequency, were used to estimate the illness management construct. Diabetes management was antecedent to metabolic control so the relation between PMDC and metabolic control was partially mediated by diabetes management. Sociodemographic variables that were statistically associated with metabolic control were added as potential confounder variables. This was a recursive model, with the PMDC having indirect effects on metabolic control through diabetes management.

Back to Article Outline

Results 

Reliability 

Means and standard deviations for the PMDC items are presented in Table 2, arranged by subdomain. All but four items (3, 5, 6, and 11) exhibited significant negative skew, indicating that parents reported high levels of monitoring of their adolescent’s diabetes management behavior. Additional statistics including item-to-total correlations (r-T) and correlations of each item with the subdomain composite score (r-C) were also computed. The item-to-total scores ranged for .05 to .68 and the item-to-composite scores ranged from .52 to .94. Internal consistency reliability of the PMDC total scale using Cronbach’s alpha was .81. Test–retest reliability using ICC was .80, indicating that the measure had a high degree of stability over a 2-week time interval.

Table 2. Correlations of items with composites (r-C) and with total PMDC scale (r-T), item means and standard deviations, and scaled factor loadings (L)
CompositeStatistic
Itemr-Cr-TMSDLa
Supervision of the Availability of Medical Supplies/Devices
11. How often do you check your child’s insulin vials to see if the expected amount has been used?.68.253.421.62.48
15. When your child’s blood glucose meter breaks or gets lost, how quickly do you know?.52.274.480.91.47
16. When your child runs out of strips and lancets, how quickly do you know?.66.404.710.691.0
17. When your child runs out of insulin, how quickly do you know?.53.184.821.19.62
18. When your child’s insulin is going to expire, how quickly do you know?.60.224.491.19.51
Monitoring Blood Glucose Checking
5. How often do you look at the readings in the BG meter?.93.612.831.46.89
6. How often do you review BG test readings with your child?.94.682.781.541.0
Oversight of Diet
12. How often do you eat meals with your child?.64.403.831.021.0
13. When you do not eat meals with your child, how often do you know what they ate?.58.333.890.89.57
14. How often does your child eat food outside of home?.72.054.141.12.39
Monitoring of Nonaherence
3. When your child skips a BG test, how often do you know?.80.523.880.97.87
4. When your child skips a BG test, how quickly do you know?.75.423.971.191.0
9. When your child skips their insulin, how often do you know?.70.294.201.01.34
10. When your child skips their insulin, how quickly do you know?.78.434.420.93.57
Direct Oversight of Diabetes Management Behaviors
1. How often were you present at home when your child tested their BG?.80.484.521.04.49
2. How often were you present in the room when your child tested their BG?.89.643.911.291.0
7. How often were you at home when your child gave themselves insulin?.76.444.670.70.25
8. How often were you present in the same room when your child giving themselves insulin?.83.513.921.39.76

Note: All loadings were significant (p < .05) except those in italics.

aLoadings were scaled by fixing the indicator with the largest factor loading to 1. Fixed loading are not tested for significance but are presumed to be significant if other loading on the same factor are.

Construct validity 

The factor structure of the PMDC was evaluated with a second-order CFA with the five subdomains constituting the first-order factors and with the general PMDC factor as the sole second-order factor. Identification was achieved by setting one path coefficient/loading on each factor, including the second-order factor, to unity. To aid in interpretation, the fixed coefficient on each factor was reassigned to the path with the largest factor loading to serve as a reference point. The fit of this initial model was unacceptable, χ2 (282.621)/df (132) = 2.14, Bollen-Stine χ2 = 174.81, p = .037, comparative fit index (CFI) = .76, and root-mean-square error of approximation (RMSEA) = .11. Modification indices were used to identify the sources of misfit. Six correlations among error terms were added. The resulting CFA indicated an acceptable fit to the data, χ2 (181.65)/df (126) = 1.44, Bollen-Stine χ2 = 165.03, p = .32, CFI = .91, RMSEA = .07. The scaled factor loadings are shown in Table 2. The four italicized loadings were not significant but retained because of their apparent conceptual relevance. Clark and Watson [29] have highlighted the importance of retaining items until instruments can be evaluated across diverse samples (e.g., clinical and nonclinical), particularly if an item appears to assess construct-relevant information, as items may demonstrate different response distributions in different populations and may have increased predictive validity in other subpopulations.

In addition, each of the paths from the general PMDC factor to the subdomain factors was significant (p < .05). The path to “Monitoring Blood Glucose Checking” was associated with the largest unstandardized coefficient and consequently was fixed to 1. Scaled path coefficients for the other subdomains were “Direct Oversight” (.55), “Monitoring of Nonadherence” (.49), “Oversight of Diet” (.23), and “Supervision of the Availability of Medical Supplies/Devices” (.10).

Concurrent validity 

Because prior studies have shown that sociodemographic variables can affect parental monitoring, the relationship between parent’s work status, number of parents in the home, race/ethnicity, and level of monitoring was examined (Table 3). As the sample was comprised predominantly of African American and white caregivers, caregivers from other minority groups were grouped with African American caregivers. Families were coded as two-parent families if the youth resided with two biological parents, a biological parent and a step-parent, or a biological parent living with a partner. Single- and two-parent families did not differ significantly in degree of monitoring of the diabetes regimen, nor did working and nonworking families, although means were in the expected direction (e.g., lower monitoring for single parents and working parents than for two-parent families and nonworking parents). However, minority parents reported significantly lower levels of oversight of adolescents’ diabetes management than nonminorities (t = 2.09, p < .04).

Table 3. Descriptive statistics for parental monitoring of diabetes care (PMDC) scale by caregiver ethnicity, working status and number of parents in the home
PMDC scoresM (SD)N
Total sample72.87(9.83)99
Caregiver ethnicitya
White75.05(9.01)46
Otherb70.98(10.21)53
Caregiver working statusc
Works outside home72.15(10.15)69
Does not work75.32(8.80)28
Number of parents in home
Two parents73.13(9.62)64
Single parent72.40(10.34)35

ap < .05.

bOther ethnicity included African American, Latino, Asian, and biracial caregivers.

cTwo caregivers did not report their work status.

Concurrent validity of the PMDC was next examined using SEM. Items making up each of the subdomains were summed and five subscales were used as composite indicators for the latent PMDC construct. The three measures of diabetes management (parent and adolescent DMS and frequency of BG checking) were used as indicators of the latent diabetes management construct and HbA1c was used as the measure of metabolic control. Because ethnicity was found to be related to scores on the PMDC, it was included in the model as a possible confounder (see Figure 1). Parameters and standard errors were estimated using the bias corrected bootstrap method in AMOS 7.0 software. The initial model had a marginal fit to the data, χ2 (46.35)/df (31) = 1.50, Bollen-Stine adjusted χ2 = 33.05, p = .075; CFI = .92, RMSEA = .07. Examination of modification indices revealed a high covariance between error terms for “Direct Oversight,” and “Oversight of Diet.” With this covariance allowed, model fit was much improved, χ2 (33.61)/df (30) = 1.12, Bollen-Stine adjusted χ2 = 32.27, p = .42, CFI = .98, RMSEA = .035. Two paths in this model were not significant, and subsequently were dropped. Model fit remained good, χ2 (36.75)/df (32) = 1.148, Bollen-Stine adjusted χ2 = 34.69, p = .41, CFI = .98, RMSEA = .035, and all path coefficients were significant (p < .05). This model shows a strong direct effect of parental monitoring as measured by the PMDC on diabetes management and an indirect effect of parental monitoring on HbA1c through diabetes management. Parental monitoring accounted for 38% of the variance in diabetes management. In addition, monitoring had a significant indirect effect on HbA1c through diabetes management (p = .007); the estimated effect was −.24.

  • View full-size image.
  • Figure 1. 

    Final SEM model results showing standardized path coefficients of relationship between parental monitoring of diabetes care (PMDC), illness management, and metabolic control (HbA1c). All paths are significant, p < .05.

Back to Article Outline

Discussion 

The purpose of the present study was to gather preliminary evidence for the reliability and validity of the parent-report version of the PMDC, an illness-specific measure of parental monitoring of adolescent illness management. Currently, there is a lack of measures in the chronic illness literature that directly assess parental supervision and oversight of adolescents’ completion of their medical care.

Results of the present study support the reliability of the PMDC. The measure had good internal consistency and was relatively stable over a 2-week interval. Study results also supported the construct validity of the PMDC. CFA allowed the identification of five domain subscales, all of which were related to the broader monitoring construct. These findings demonstrate that parent behaviors such as monitoring the availability of diabetes supplies and medical devices, reviewing blood glucose meter readings, oversight of adolescents’ diet, awareness of the occurrence of nonadherence behavior such as skipping shots, and being present at home when diabetes care is completed are different means by which parents can ensure that the adolescent completes their diabetes care. Subscales that evaluated whether or not the parent was present during diabetes care completion and whether the parent monitored the adolescent’s blood glucose checking through review of the blood glucose meter were strongly related to the monitoring construct. In light of the prior studies that highlight the importance of parental involvement in the youth’s diabetes regimen for prompting good metabolic control [12], [30] and the frequent recommendation in cases of poor illness management that the caregiver directly oversee completion of the regimen [31], [32], this finding also supports the validity of the measure. Furthermore, results of structural equation modeling showed that the PMDC was strongly related to illness management and, through illness management, to metabolic control. Parental monitoring accounted for 38% of the variance in diabetes management. Parental monitoring also had a significant indirect effect on HbA1c, indicating that close parental supervision and oversight of the diabetes regimen predicted better health outcomes in adolescents with diabetes. Additional studies will be needed to evaluate whether the PMDC demonstrates measurement invariance for older adolescents, as prior studies have suggested that parents may need to adjust their oversight of and involvement with diabetes management to account for the developmental needs of older youth [33]. Testing of a youth-report version of the PMDC to obtain youth perspectives on parental monitoring would also help addresses such questions.

The finding that items evaluating direct supervision of diabetes care activities were strongly related to the parental monitoring construct—and hence to adolescent illness and metabolic control—also addresses a current debate in the literature on parental monitoring. Stattin and Kerr [34] have suggested that because the majority of instruments evaluating parental monitoring in fact ask about parental knowledge of youth activities and whereabouts, rather than about parental tracking and surveillance of their child, measures of parental monitoring may, in fact, assess youths’ disclosure of their daily activities. They argue that positive parent–child communication, rather than parental attempts to independently gather information about youth activities, is the crucial ingredient for decreasing risky behaviors. In fact, their data indicated that youths’ spontaneous disclosure of information about their activities was more explanatory of youth adjustment and risky behavior than parental attempts to gather information about what the youth had done. However, items from the PMDC that measured parental “knowledge” of whether the adolescent had engaged in noncompliant behavior such as skipping insulin doses or blood glucose checks were not the most highly related to overall monitoring in the present study. Therefore, results of the present study suggest that at least for diabetes management, direct supervision, and surveillance of adolescents’ daily diabetes care activities is an important component of predicting good health outcomes, and that it should not be dismissed in favor of simply asking the youth about his or her completion of diabetes care tasks.

Although single-parent families have shown to have lower levels of parental monitoring in the broader adolescent risk literature [1], [2], results of the present study did not suggest that single parents provided less oversight of the diabetes regimen than parents in two-parent families. Because only 30% of the current sample of 100 families were single parents, power to detect such differences may have been low. However, minority parents did report having lower levels of diabetes-specific monitoring than white parents. This may have occurred due a pooling of other risk factors that are known to negatively affect monitoring, such as social disadvantage, among minority families [17], [35]. Alternatively, lower levels of monitoring in these families could reflect differences in beliefs in the value of early emancipation of adolescents [36] or other factors.

Limitations of the present study include a small sample size that precluded the investigation of potential moderator variables. For example, it is possible that the relative importance of the different PMDC factors may be related to sample sociodemographics. Small sample size was also a factor when CFA was conducted with the PMDC. Several method factors due to common phrases across items, for example, “how often do you know,” and serial dependencies across items involving parental presence in the home and presence in the room were identified. Aggregating items to the subscale level as was done when concurrent validity was evaluated effectively eliminated these sources of method variance; however, these issues should be considered in further psychometric development of the scale when larger sample size may allow better identification of items that should be eliminated or reworded due to such concerns. Construct validity data presented here is also preliminary, and replication of initial findings regarding PMDC factors with larger samples is warranted.

Although the sample was reasonably diverse and approximately half of the sample was white, the sample contained more minority and single-parent youth than are represented in the general population of youth with type 1 diabetes in the United States. In addition, due to the number of minority youth in the sample, and the well-known connection between minority statues and metabolic control [37], [38], average metabolic control was relatively poor. It will be important to determine if all PMDC items perform equally well in other populations of youth with diabetes, including those whose diabetes is better controlled and those with differing demographic compositions. Nonetheless, the measure addresses an important gap in the instrumentation literature for youth with diabetes that has historically used white, higher socioeconomic status samples.

In addition, the study was cross-sectional in nature, and hence cannot address important aspects of instrument validity such as the PMDC’s predictive validity. Longitudinal studies of illness-specific monitoring could also help to address the interaction of youth development stage with monitoring to identify when and how parental monitoring should optimally be decreased to promote the development of independent self-care skills. In addition, longitudinal studies could help to clarify the processes by which parental monitoring and youth illness management influence one another over time. Finally, it should be acknowledged that as is often true of research on chronically ill children, the parents who participated in the present study were predominantly female, and therefore the psychometrics properties of the PMDC when used with fathers requires further evaluation.

The PMDC represents an important first step in the development of illness-specific measures of parental monitoring for use with adolescents with chronic medical conditions. Additional work in this area will promote assessment of the ways in which instrumental parenting behaviors, such as monitoring, contribute to the maintenance of good adherence and health outcomes for adolescents, as well as fostering the development of interventions to promote effective parental monitoring.

Back to Article Outline

References 

  1. French DC, Dishion T. Predictors of early initiation of sexual intercourse among high-risk adolescents. J Early Adolesc. 2003;23:295–315
  2. Rodgers KB, Rose HA. Personal, family, and school factors related to adolescent academic performance: a comparison by family structure. Marr Fam Rev. 2001;33:47–61
  3. Chilcoat HD, Anthony JC. Impact of parent monitoring on initiation of drug use through late childhood. J Am Acad Child Adolesc Psychiatry. 1996;35:91–100
  4. Dishion TJ, McMahon RJ. Parental monitoring and the prevention of child and adolescent problem behavior: a conceptual and empirical formulation. Clin Child Fam Psychol Rev. 1998;1:61–75
  5. Li X, Stanton B, Feigelman S. Impact of perceived parental monitoring on adolescent risk behavior over 4 years. J Adolesc Health. 2000;27:49–56
  6. Griffin KW, Botvin GJ, Scheier LM, et al. Parenting practices as predictors of substance use, delinquency, and aggression among urban minority youth: moderating effects of family structure and gender. Psychol Addict Behav. 2000;14:174–184
  7. Pettit GS, Bates JE, Dodge Ka, Meece DW. The impact of after-school peer contact on early adolescent externalizing problems is moderated by parental monitoring, perceived neighborhood safety, and prior adjustment. Child Dev. 1999;70:768–778
  8. Dishion TJ, Nelson SE, Kavanagh K. The family check-up with high-risk young adolescents: preventing early onset substsance use by parent monitoring. Behav Ther. 2003;34:553–571
  9. Stanton B, Cole M, Galbraith J, et al. Randomized trial of a parent intervention: parents can make a difference in long-term adolescent risk behaviors, perceptions, and knowledge. Arch Pediatr Adolesc Med. 2004;158:947–955
  10. La Greca AM, Follansbee D, Skyler JS. Developmental and behavioral aspects of diabetes management in youngsters. Child Health Care. 1990;19:132–139
  11. Mellins CA, Brackis-Cott E, Dolezal C, Abrams EJ. The role of psychosocial and family factors in adherence to antiretroviral treatment in human immunodeficiency virus-infected children. Pediatr Infect Dis. 2004;23:1035–1041
  12. Wysocki T, Taylor A, Hough BS, et al. Deviation from developmentally appropriate self-care autonomy: association with diabetes outcome. Diabetes Care. 1996;19:119–125
  13. Shaw RJ, Palmer L, Balsey C. A typology of non-adherence in pediatric renal transplant recipients. Pediatr Transplant. 2003;7:489–493
  14. Anderson BJ, Auslander WF, Jung KC, et al. Assessing family sharing of diabetes responsibilities. J Pediatr Psychol. 1990;15:477–492
  15. Palmer D, Berg A, Wiebe D, et al. The role of autonomy and pubertal status in understanding age differences in maternal involvement in diabetes responsibility across adolescence. J Pediatr Psychol. 2004;29:35–46
  16. Walders N, Drotar D, Kercsmar C. The allocation of family responsibility for asthma management tasks in African-American adolescents. J Asthma. 2000;37:89–99
  17. Chilcoat HD, Breslau N, Anthony JC. Potential barriers to parent monitoring: social disadvantage, marital status, and maternal psychiatric disorder. J Am Acad Child Adolesc Psychiatry. 1996;35:1673–1682
  18. Patterson GR, Stouthamer-Loeber M. The correlation of family mangement practices and delinquency. Child Dev. 1984;55:1299–1307
  19. Steinberg L, Lamborn SD, Dornbusch SM, Darling N. Impact of parenting practices on adolescent achievement: authoritative parenting, school involvement, and ecouragement to succeed. Child Dev. 1992;63:1266–1291
  20. McQuaid EL, Kopel SJ, Klein RB, Fritz GK. Medication adherence in pediatric asthma: reasoning, responsibility, and behavior. J Pediatr Psychol. 2003;28:323–333
  21. Wysocki T. Associations among teen–parent relationships, metabolic control, and adjustment to diabetes in adolescents. J Pediactr Psychol. 1993;18:441–452
  22. Frey M, Ellis D, Naar-King S, Greger N. Diabetes management in adolescents in poor metabolic control. Diabetes Educ. 2004;30:647–657
  23. Frey MA, Denyes MJ. Health and illness self-care in adolescents with IDDM: a test of Orem’s theory. Adv Nurs Sci. 1989;12:67–75
  24. Schilling LS, Grey M, Knafl K. A review of measures of self-management of type 1 diabetes by youth and their parents. Diabetes Educ. 2002;28:796–808
  25. Arbuckle JL. Amos (Version 7.0). [Computer Program] Chicago: SPSS; 2006;
  26. Shrout PE, Bolger N. Meditation in experimental and non experimental studies: new procedures and recommendations. Psychol Methods. 2002;7:422–445
  27. Bollen KA, Stine R. Direct and indirect effects: classical and bootstrap estimates of variability. Sociol Methodol. 1990;20:115–140
  28. Kishton JM, Widaman KF. Unidimensional versus domain representative parceling of questionnaire items: an empirical example. Educ Psychol Measure. 1994;54:757–765
  29. Clark LA, Watson D. Constructing validity: basic issues in objective scale development. Psychol Assess. 1995;7:309–319
  30. Anderson RM, Barr PA, Edwards GJ, et al. Using focus groups to identify psychosocial issues of urban black individuals with diabetes. Diabetes Educ. 1996;22:28–33
  31. Golden MP, Herrold AJ, Orr DP. An approach to prevention of recurrent diabetic ketoacidosis in the pediatric population. J Pediatr. 1985;107:195–200
  32. Gray DL, Marrero DG, Godfrey C, et al. Chronic poor metabolic control in the pediatric population: a stepwise intervention program. Diabetes Educ. 1988;14:516–520
  33. Wiebe DJ, Berg CA, Korbel C, et al. Children’s appraisals of maternal involvement in coping with diabetes: enhancing our understanding of adherence, metabolic control, and quality of life across adolescence. J Pediatr Psychol. 2005;30:167–178
  34. Stattin H, Kerr M. Parental monitoring: a reinterpretation. Child Dev. 2000;71:1072–1082
  35. Conger RD, Wallace LE, Sun Y, et al. Economic pressure in African American families: a replication and extension of the family stress model. Dev Psychol. 2002;38:179–193
  36. Roche KM, Ensminger ME, Chilcoat H, Storr C. Establishing independence in low-income urban areas: the relationship to adolescent aggressive behavior. J Marr Fam. 2003;65:668–680
  37. Delamater AM, Shaw KH, Applegate EB, et al. Risk for metabolic control problems in minority youth with diabetes. Diabetes Care. 1999;22:700–705
  38. Frey MA, Templin T, Ellis DA, Gutai J. Metabolic control in the first 3 years post-diagnosis of type 1 diabetes in African American and European American youths. Pediatr Diabetes. 2007;8:220–227

PII: S1054-139X(07)00342-4

doi:10.1016/j.jadohealth.2007.08.012

Journal of Adolescent Health
Volume 42, Issue 2 , Pages 146-153, February 2008

Article Tools

* Image Usage & Resolution