The effects of a spouse implemented contingency contract on asthma medication adherence.
This study investigated the effects of a contingency contract on
asthma medication adherence. Participants were a 30 year-old woman with
asthma and her husband. The researcher initially implemented the
contingency contract and later trained the husband to implement the same
contract. An ABAC reversal design showed that medication adherence
increased over baseline levels during the contingency contract
conditions, whether implemented by researcher or husband. Pulmonary
functions increased with medication adherence with a correlation of .86.
Follow-up observations showed the program survived with spouse
implementation nine months post-research suggesting that use of a
natural supervisor may promote sustainability.
Keywords: asthma, adherence, contingency contract, program survival, sustainability, natural supervisor
Hillman, Heidi L.
Miller, L. Keith
|Publication:||Name: The Behavior Analyst Today Publisher: Behavior Analyst Online Audience: Academic Format: Magazine/Journal Subject: Psychology and mental health Copyright: COPYRIGHT 2009 Behavior Analyst Online ISSN: 1539-4352|
|Issue:||Date: Wntr, 2009 Source Volume: 10 Source Issue: 1|
|Topic:||Event Code: 490 Contracts & orders let; 610 Contracts & orders received Computer Subject: Contract agreement|
|Product:||Product Code: 2834481 Antiasthmatic Preparations NAICS Code: 325412 Pharmaceutical Preparation Manufacturing SIC Code: 2834 Pharmaceutical preparations|
Asthma prevalence, morbidity, and mortality in the United States
have increased during the past decade despite major national efforts to
improve asthma awareness and care (Centers for Disease Control and
Prevention, 1998). One cause of this increase may be asthma medication
non adherence. Asthma sufferers' adherence to asthma medications is
very low. Asthma sufferers are less likely to adhere to their medication
regimen if the asthma symptoms are either mild or severe, if the
medication is expensive, and if the medication side-effects are aversive
(Cramer & Spilker, 1991). Several studies have compared
participants' self-report with direct measures (e.g., urine
testing, pill counting) of medication adherence, and overall these
studies suggest that adherence ranges between 10% and 60% (Spector,
Applied behavior analysts have proposed various factors for improving adherence to treatment regimens and have repeatedly shown that adherence responds to properly managed contingencies (Allen & Warzack, 2000). Contingency management systems have been used to increase adherence in a variety of settings, such as schools, communities, and medical settings. Da Costa, Rapoff, Lemanek, and Goldstein (1997) provide one example where a parent effectively implemented a contingency management system to increase asthma medication adherence of two children. The intervention included asthma education and a token system. The token system involved the children earning points for medication adherence and exchanging the points for privileges. Failure to take the prescribed medications resulted in a loss of privileges for one day. However, a spouse-implemented contract on asthma medication adherence by an adult spouse has not been analyzed.
Hence, the present study is an extension of da Costa et al. (1997) in that similar techniques (e.g. contingency contract, pulmonary readings) were employed, except the participant dyad was a husband and wife instead of a parent and child. The purpose of this study was to (1) analyze the effectiveness of a contingency contract on asthma medication adherence, (2) to evaluate the effect of transferring contract implementation to the husband and (3) to observe maintenance during nine months of follow-up probes. In addition, the study measured the participant's pulmonary functioning to assess the effect of asthma medication adherence.
Participants and Setting
The researcher used an electronic monitor to observe Tracy's adherence to a medically prescribed regimen. The regimen prescribed inhaling one dose of Serevent every morning and another dose every evening. The researcher measured adherence by an electronic chronology monitor called a Doser[TM] that attached to the Serevent canister and recorded the number and date of each actuation. Doser-recorded actuations could be displayed on an LCD panel, and stored in memory. The researcher considered Doser readings to be reliable when Tracy's, George's and the researcher's number of Doser recorded actuations matched. For example, the researcher, Tracy and George would individually write down the number located on the LCD display and the three numbers would be compared. Reliability was 100% for each condition.
In addition, the researcher used a peak flow meter to determine the effect of medication adherence on pulmonary functioning. Tracy's peak flow was assessed weekly using a hand held peak flow meter, a device that measures the peak volume of air that Tracey exhaled (in Liters per minute). This measure was taken because one of the symptoms of asthma is an obstructed airway leading to impaired peak flow. The researcher followed the standard protocol of instructing Tracy to exhale as forcibly as she could into the peak flow meter with three separate breaths while standing and then charting the highest of the three readings (American Lung Association, 2008).
A mutually agreed upon contract stipulated that Tracy increase her weekly Serevent use by at least one dose over the previous week, or use Serevent twice daily for seven consecutive days. The researcher calculated medication adherence at each weekly home visit by writing down the actuation number shown on the Doser LCD panel and subtracting the previous week's Doser reading from it. The researcher, Tracy and George compared the Doser recorded actuation number with the previous week to decide if Tracy met the weekly criterion. Each week that Tracy met her criterion, she blindly drew a recreational activity (e.g., go to a movie with George, shop at a specific store) from a mutually agreed upon pool of activities. If Tracy did not like the first pick, she put the paper aside and drew another activity. The activity was completed on Saturday with George. The researcher called each Sunday to check if Tracy received her reward. If Tracy did not meet her criterion, she did not draw an activity.
The researcher implemented the contract during the contract-with-researcher condition and also trained George to implement the contract. George then implemented the contract during the contract-with-spouse condition. The researcher observed the fidelity with which George implemented the contingency contract weekly during the spouse-implemented contract condition by observing how many of the steps mentioned above (e.g., recording Doser number, determining if Tracy met the criterion) George correctly performed. George correctly implemented the contract 100% of the times observed.
The design consisted of a two-week baseline with no contract, a six-week researcher implemented contingency contract, returning to baseline for five weeks, a seven-week spouse implemented contract, and finally five follow-up probes during nine months with spouse implementation of the contract. The experiment used an ABAC reversal design.
[FIGURE 1 OMITTED]
Figure 1 shows Tracy's weekly asthma medication adherence and peak flow. During baseline, Tracy's medication adherence was six out of a maximum of 14. During the researcher-implemented contract, Tracy's adherence increased to an average of 10.5 doses per week and stabilized at 11.8 doses. During reversal to baseline, Tracy's adherence decreased to an average of nine doses per week and stabilized at 8 doses. During the spouse-implemented contract, medication adherence returned to an average of 11.8 doses per week and stabilized at 12.8 doses. The follow-up probes showed that Tracy's asthma medication averaged 11 doses per week. Tracy's asthma medication adherence was higher during contingency contract phases and follow-up than during baseline conditions.
Figure 1 also shows Tracy's weekly peak flow reading. In general the level of Peak flow followed the level of medication adherence with a delay of about one week. The correlation between medication adherence and peak flow (delayed by one week) was .86--accounting for over 70% of the variance in peak flow.
The fact that Tracy's medication adherence repeatedly increased with the introduction of the contract and decreased with its withdrawal suggests that the contract and not other factors caused the increase in medication adherence. Further, the high correlation between peak flow and medication adherence suggests that adherence did strongly influence pulmonary functioning.
The feasibility of spousal implementation of the contract is suggested by the fact that Tracy's rate of adherence was comparable when George implemented the contract as when the researcher implemented it. George's implementation showed perfect fidelity with the procedure with the researcher present during the "spouse-implemented condition". However, the fidelity showed a moderate decrease during the "follow-up" condition when the researcher was present for only about one in eight implementations.
George and Tracy made two modifications that reduced fidelity during follow-up. The first modification actually made the contingency stricter. Tracy decided she had to achieve at least 11 doses during the week to obtain a reinforcing activity. The second modification was that Tracy directly chose her preferred activity instead of blindly drawing it.
Many behavior analysts argue that most interventions fail to survive after withdrawal of direct research supervision (e.g., Welsh, Miller & Altus, 1994). The robust survival of the contract in the present study invites speculation. George frequently mentioned better symptom-management leading to expanded recreational activities with Tracy as a desirable outcome of implementing the contract. The possibility that George's implementation of the contract was reinforced by those outcomes is suggested by the survival of the program during the nine-month follow-up period. Thus transferring supervision to a person whose supervisory behavior may be differentially reinforced by the outcomes of the program may be one strategy for promoting program survival. Direct experimental analysis of the "natural supervisor" strategy may be worth pursuing in future research.
The current study has several limitations. First, the study used only one person with asthma and her husband. We, therefore, cannot conclude these results will generalize to other people with low adherence to asthma medication regimens. Additional research is needed using more participants. Second, the study observed actuations of the Doser rather than directly observing inhalations. However, weekly measurements of Tracy's peak flow showed increases during treatment and decreases during reversals that closely parallel the variations in Doser readings. Therefore we can have reasonable confidence that Doser actuations are a valid measurement of inhalations. Third, it is possible that reactivity to the presence of the researcher may explain the continued use of the procedure during follow up. However, because Doser readings are cumulative it was possible to obtain an unobtrusive measure of doses during weeks when the researcher was not scheduled to observe. The average doses pre week were about the same whether the researcher was present or not--suggesting that the spouse did continue to implement the contract. Never-the-less, future research should examine the reactivity effects of the researcher on the survival of an intervention.
The study used an ABAC reversal design because of its high level of internal validity--ruling out most extraneous factors. One reason for using the reversal design was the interest in studying how adherence behavior changed in the presence of a contingency contract and to determine if treatment effects persisted after withdrawal of the contingency contract. The advantage of reversal designs is they help rule out the interpretation that some variable other than the intervention may be responsible for change in the behavior; a convincing case can then be made that the intervention is effective. Reversal designs, however, are not without their disadvantages--primarily ethical issues concerning withdrawal of the intervention.
In the case of this study, to minimize the ethical concern of withdrawing the contingency contract after increased medication adherence and increased pulmonary functioning were noted, Tracy was informed of the withdrawal and asked if she would be willing to discontinue using the contingency contract; she agreed. After the withdrawal of the contingency contract Tracy's medication adherence and pulmonary functioning decreased; however at the lowest datum point in the return to no contract condition medication adherence and pulmonary functioning was greater than during the first no contract condition.
In summary the present study showed that a contingency contract implemented by a researcher could increase adherence to an asthma medication regimen. The study also showed that a spouse could implement the same program with similar results. Finally, the data suggest that the use of a "natural supervisor"--such as a spouse--may promote the post-research survival of behavioral programs.
Allen, K. D., & Warzak, W. J. (2000). The problem of parental non-adherence in clinical behavior analysis: Effective treatment is not enough. Journal of Applied Behavior Analysis, 33, 373-391.
American Lung Association, (2008, June) Peak Flow Meters. Retrieved July 29, 2008, from http://www.lungusa.org/site/pp.asp?c=dvLUK9O0E&b=22586#6
Centers for Disease Control and Prevention. (1998). Surveillance for asthma: United States, 1960-1995. Morbidity and Mortality Report, 47, SS-1, 1-27.
Cramer, J., & Spilker, B. (1991). Patient compliance in medical practice and clinical trials. New York: Raven Press.
da Costa, I., G., Rapoff, M., A., Lemanek, K., & Goldstein, G., L. (1997). Improving adherence to education regimens for children with asthma and its effects on clinical outcome. Journal of Applied Behavior Analysis, 30, 687-691.
Spector, S. (2000). Noncompliance with asthma therapy: Are there solutions? Journal of Asthma, 37, 381-388.
Welsh, T. M., Miller, L. K., & Altus, D. E. (1994). Programming for survival: A meetings system that survives 8 years later. Journal of Applied Behavior Analysis, 9, 423-433.
L. Keith Miller
|Gale Copyright:||Copyright 2009 Gale, Cengage Learning. All rights reserved.|