Previous research on the impact of social bonds upon substance use has been mostly cross-sectional. A new study investigated the longitudinal impacts of social bonds on nonmedical prescription drug use (NMPDU). Data were obtained from the National Longitudinal Study of Adolescent to Adult Health, which contained four waves of data collection. Participants were aged 14-16 years at Wave 1 and 28-30 years at Wave 4. At each wave, they reported their NMPDU since the previous wave. Social bond measurements included the marital bond, parent-child bond, education bond, religious bond, and several others. Participants were grouped into trajectories based on their responses for each bond at each wave. The authors used group-based latent trajectory modeling as the main analytical tool, as well as multiple imputation technique to handle missing data. Results showed 5.8% of the sample reported NMPDU at baseline and 18.6% reported NMPDU at Wave 4. Overall, self-reported relationship with parents declined in quality at Wave 3, but rose back to baseline levels at Wave 4. Most participants were enrolled in an educational institution at Waves 1 and 2. There were three trajectories for religious bond: 34% of the sample consistently attended church with high frequency, 43.2% gradually decreased attendance (“high decrease” group), and 22.8% never really frequented church, but increased attendance slowly over time (“low increase” group). Model 1 used “constantly high level” trajectories as the reference group. It showed the “low increase” and “high decrease” religious bond groups were more likely to engage in NMPDU (both ps < 0.001). Similar trajectories were created for other bonds. The “low increase” trajectory of familial bond had elevated risk of NMPDU at Wave 4 (p< 0.001), but such risk was not observed in the “high decrease” trajectory. Delayed marriage was found to have a protective impact on likelihood of NMPDU, while enrollment in higher education or training may be a possible risk factor. Baseline NMPDU was associated with a significantly higher likelihood of NMPDU in at Wave 4 (p < 0.001). In Model 2 (which used “low initial” as the reference group) for most types of social bond, people in the “constant high” trajectories were significantly less likely to report NMPDU than their counterparts in all other trajectories. This implies cumulative level of social bond is important and the initial bonding strength may not matter as much as the contemporaneous bonding strength.
Take away: In this nationally representative sample, people who belonged in a constant and high-level trajectory of social bond were significantly less likely to report nonmedical use of prescription drugs in their late twenties than all others.